prometheus/storage/fanout.go

874 lines
24 KiB
Go
Raw Normal View History

2017-05-10 02:44:13 -07:00
// Copyright 2017 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
2017-05-10 02:44:13 -07:00
package storage
import (
"bytes"
2017-05-10 02:44:13 -07:00
"container/heap"
"context"
"sort"
2017-05-10 02:44:13 -07:00
"strings"
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
"sync"
2017-05-10 02:44:13 -07:00
2017-08-11 11:45:52 -07:00
"github.com/go-kit/kit/log"
"github.com/go-kit/kit/log/level"
"github.com/pkg/errors"
"github.com/prometheus/common/model"
2017-05-10 02:44:13 -07:00
"github.com/prometheus/prometheus/pkg/labels"
"github.com/prometheus/prometheus/tsdb/chunkenc"
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
"github.com/prometheus/prometheus/tsdb/chunks"
tsdb_errors "github.com/prometheus/prometheus/tsdb/errors"
2017-05-10 02:44:13 -07:00
)
type fanout struct {
2017-08-11 11:45:52 -07:00
logger log.Logger
primary Storage
secondaries []Storage
2017-05-10 02:44:13 -07:00
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// NewFanout returns a new fanout Storage, which proxies reads and writes
2017-05-10 02:44:13 -07:00
// through to multiple underlying storages.
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
//
// The difference between primary and secondary Storage is only for read (Querier) path and it goes as follows:
// * If the primary querier returns an error, then any of the Querier operations will fail.
// * If any secondary querier returns an error the result from that queries is discarded. The overall operation will succeed,
// and the error from the secondary querier will be returned as a warning.
//
// NOTE: In the case of Prometheus, it treats all remote storages as secondary / best effort.
2017-08-11 11:45:52 -07:00
func NewFanout(logger log.Logger, primary Storage, secondaries ...Storage) Storage {
2017-05-10 02:44:13 -07:00
return &fanout{
2017-08-11 11:45:52 -07:00
logger: logger,
primary: primary,
secondaries: secondaries,
2017-05-10 02:44:13 -07:00
}
}
// StartTime implements the Storage interface.
func (f *fanout) StartTime() (int64, error) {
// StartTime of a fanout should be the earliest StartTime of all its storages,
// both primary and secondaries.
firstTime, err := f.primary.StartTime()
if err != nil {
return int64(model.Latest), err
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
for _, s := range f.secondaries {
t, err := s.StartTime()
if err != nil {
return int64(model.Latest), err
}
if t < firstTime {
firstTime = t
}
}
return firstTime, nil
}
func (f *fanout) Querier(ctx context.Context, mint, maxt int64) (Querier, error) {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
primary, err := f.primary.Querier(ctx, mint, maxt)
if err != nil {
return nil, err
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
secondaries := make([]Querier, 0, len(f.secondaries))
for _, storage := range f.secondaries {
querier, err := storage.Querier(ctx, mint, maxt)
2017-05-10 02:44:13 -07:00
if err != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// Close already open Queriers, append potential errors to returned error.
errs := tsdb_errors.MultiError{err}
errs.Add(primary.Close())
for _, q := range secondaries {
errs.Add(q.Close())
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
return nil, errs.Err()
2017-05-10 02:44:13 -07:00
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
secondaries = append(secondaries, querier)
}
return NewMergeQuerier(primary, secondaries, ChainedSeriesMerge), nil
2017-05-10 02:44:13 -07:00
}
func (f *fanout) ChunkQuerier(ctx context.Context, mint, maxt int64) (ChunkQuerier, error) {
primary, err := f.primary.ChunkQuerier(ctx, mint, maxt)
if err != nil {
return nil, err
}
secondaries := make([]ChunkQuerier, 0, len(f.secondaries))
for _, storage := range f.secondaries {
querier, err := storage.ChunkQuerier(ctx, mint, maxt)
if err != nil {
// Close already open Queriers, append potential errors to returned error.
errs := tsdb_errors.MultiError{err}
errs.Add(primary.Close())
for _, q := range secondaries {
errs.Add(q.Close())
}
return nil, errs.Err()
}
secondaries = append(secondaries, querier)
}
return NewMergeChunkQuerier(primary, secondaries, NewCompactingChunkSeriesMerger(ChainedSeriesMerge)), nil
}
func (f *fanout) Appender() Appender {
primary := f.primary.Appender()
secondaries := make([]Appender, 0, len(f.secondaries))
for _, storage := range f.secondaries {
secondaries = append(secondaries, storage.Appender())
2017-05-10 02:44:13 -07:00
}
return &fanoutAppender{
2017-08-11 11:45:52 -07:00
logger: f.logger,
primary: primary,
secondaries: secondaries,
}
2017-05-10 02:44:13 -07:00
}
// Close closes the storage and all its underlying resources.
func (f *fanout) Close() error {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
errs := tsdb_errors.MultiError{}
errs.Add(f.primary.Close())
for _, s := range f.secondaries {
errs.Add(s.Close())
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
return errs.Err()
2017-05-10 02:44:13 -07:00
}
// fanoutAppender implements Appender.
type fanoutAppender struct {
2017-08-11 11:45:52 -07:00
logger log.Logger
primary Appender
secondaries []Appender
2017-05-10 02:44:13 -07:00
}
func (f *fanoutAppender) Add(l labels.Labels, t int64, v float64) (uint64, error) {
ref, err := f.primary.Add(l, t, v)
if err != nil {
return ref, err
}
for _, appender := range f.secondaries {
2017-05-10 02:44:13 -07:00
if _, err := appender.Add(l, t, v); err != nil {
return 0, err
2017-05-10 02:44:13 -07:00
}
}
return ref, nil
2017-05-10 02:44:13 -07:00
}
func (f *fanoutAppender) AddFast(ref uint64, t int64, v float64) error {
if err := f.primary.AddFast(ref, t, v); err != nil {
return err
}
for _, appender := range f.secondaries {
if err := appender.AddFast(ref, t, v); err != nil {
return err
}
}
return nil
2017-05-10 02:44:13 -07:00
}
func (f *fanoutAppender) Commit() (err error) {
err = f.primary.Commit()
for _, appender := range f.secondaries {
if err == nil {
err = appender.Commit()
} else {
if rollbackErr := appender.Rollback(); rollbackErr != nil {
2017-08-11 11:45:52 -07:00
level.Error(f.logger).Log("msg", "Squashed rollback error on commit", "err", rollbackErr)
}
2017-05-10 02:44:13 -07:00
}
}
return
2017-05-10 02:44:13 -07:00
}
func (f *fanoutAppender) Rollback() (err error) {
err = f.primary.Rollback()
for _, appender := range f.secondaries {
rollbackErr := appender.Rollback()
if err == nil {
err = rollbackErr
} else if rollbackErr != nil {
2017-08-11 11:45:52 -07:00
level.Error(f.logger).Log("msg", "Squashed rollback error on rollback", "err", rollbackErr)
2017-05-10 02:44:13 -07:00
}
}
return nil
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
type mergeGenericQuerier struct {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
queriers []genericQuerier
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// mergeFn is used when we see series from different queriers Selects with the same labels.
mergeFn genericSeriesMergeFunc
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// NewMergeQuerier returns a new Querier that merges results of given primary and slice of secondary queriers.
// See NewFanout commentary to learn more about primary vs secondary differences.
//
// In case of overlaps between the data given by primary + secondaries Selects, merge function will be used.
func NewMergeQuerier(primary Querier, secondaries []Querier, mergeFn VerticalSeriesMergeFunc) Querier {
queriers := make([]genericQuerier, 0, len(secondaries)+1)
if primary != nil {
queriers = append(queriers, newGenericQuerierFrom(primary))
}
for _, querier := range secondaries {
if _, ok := querier.(noopQuerier); !ok && querier != nil {
queriers = append(queriers, newSecondaryQuerierFrom(querier))
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
return &querierAdapter{&mergeGenericQuerier{
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
mergeFn: (&seriesMergerAdapter{VerticalSeriesMergeFunc: mergeFn}).Merge,
queriers: queriers,
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}}
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// NewMergeChunkQuerier returns a new ChunkQuerier that merges results of given primary and slice of secondary chunk queriers.
// See NewFanout commentary to learn more about primary vs secondary differences.
//
// In case of overlaps between the data given by primary + secondaries Selects, merge function will be used.
// TODO(bwplotka): Currently merge will compact overlapping chunks with bigger chunk, without limit. Split it: https://github.com/prometheus/tsdb/issues/670
func NewMergeChunkQuerier(primary ChunkQuerier, secondaries []ChunkQuerier, mergeFn VerticalChunkSeriesMergeFunc) ChunkQuerier {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
queriers := make([]genericQuerier, 0, len(secondaries)+1)
if primary != nil {
queriers = append(queriers, newGenericQuerierFromChunk(primary))
}
for _, querier := range secondaries {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
if _, ok := querier.(noopChunkQuerier); !ok && querier != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
queriers = append(queriers, newSecondaryQuerierFromChunk(querier))
}
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return &chunkQuerierAdapter{&mergeGenericQuerier{
mergeFn: (&chunkSeriesMergerAdapter{VerticalChunkSeriesMergeFunc: mergeFn}).Merge,
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
queriers: queriers,
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}}
2017-05-10 02:44:13 -07:00
}
// Select returns a set of series that matches the given label matchers.
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
func (q *mergeGenericQuerier) Select(sortSeries bool, hints *SelectHints, matchers ...*labels.Matcher) genericSeriesSet {
if len(q.queriers) == 1 {
return q.queriers[0].Select(sortSeries, hints, matchers...)
Don't sort postings if we only have one block. Sorting the heads postings can be quite slow. We only need sorted series when merging with another querier, so only sort then. This will make big queries that only touch the head faster, though queries that touch both the head and a block will still be the same speed. This probably won't help much with graphing unless the range is under an hour, however it should make most recording rules faster. Add gaurantee that remote read streaming produces sorted series. PromQL benchmarks for histograms show only 2-3% improvement, but they're only over 1k series. benchmark old ns/op new ns/op delta BenchmarkQuerierSelect/Head/1of1000000-4 1375486282 507657736 -63.09% BenchmarkQuerierSelect/Head/10of1000000-4 1387859004 507769850 -63.41% BenchmarkQuerierSelect/Head/100of1000000-4 1387087935 506029110 -63.52% BenchmarkQuerierSelect/Head/1000of1000000-4 1386869064 504521986 -63.62% BenchmarkQuerierSelect/Head/10000of1000000-4 1386213685 505210422 -63.55% BenchmarkQuerierSelect/Head/100000of1000000-4 1392754988 529842406 -61.96% BenchmarkQuerierSelect/Head/1000000of1000000-4 1569414722 725059506 -53.80% BenchmarkQuerierSelect/SortedHead/1of1000000-4 1381019902 1370495863 -0.76% BenchmarkQuerierSelect/SortedHead/10of1000000-4 1375696209 1366789468 -0.65% BenchmarkQuerierSelect/SortedHead/100of1000000-4 1386009422 1364519297 -1.55% BenchmarkQuerierSelect/SortedHead/1000of1000000-4 1377700532 1364486191 -0.96% BenchmarkQuerierSelect/SortedHead/10000of1000000-4 1383539536 1369545314 -1.01% BenchmarkQuerierSelect/SortedHead/100000of1000000-4 1410089163 1394731339 -1.09% BenchmarkQuerierSelect/SortedHead/1000000of1000000-4 1634744148 1581554956 -3.25% BenchmarkQuerierSelect/Block/1of1000000-4 881741242 879839470 -0.22% BenchmarkQuerierSelect/Block/10of1000000-4 880381562 882846038 +0.28% BenchmarkQuerierSelect/Block/100of1000000-4 887519357 881016916 -0.73% BenchmarkQuerierSelect/Block/1000of1000000-4 902194205 883433524 -2.08% BenchmarkQuerierSelect/Block/10000of1000000-4 892321964 885130170 -0.81% BenchmarkQuerierSelect/Block/100000of1000000-4 938604466 933527150 -0.54% BenchmarkQuerierSelect/Block/1000000of1000000-4 1313510845 1295881124 -1.34% benchmark old allocs new allocs delta BenchmarkQuerierSelect/Head/1of1000000-4 4000056 4000018 -0.00% BenchmarkQuerierSelect/Head/10of1000000-4 4000074 4000036 -0.00% BenchmarkQuerierSelect/Head/100of1000000-4 4000254 4000216 -0.00% BenchmarkQuerierSelect/Head/1000of1000000-4 4002054 4002016 -0.00% BenchmarkQuerierSelect/Head/10000of1000000-4 4020054 4020016 -0.00% BenchmarkQuerierSelect/Head/100000of1000000-4 4200054 4200016 -0.00% BenchmarkQuerierSelect/Head/1000000of1000000-4 6000054 6000016 -0.00% BenchmarkQuerierSelect/SortedHead/1of1000000-4 4000071 4000071 +0.00% BenchmarkQuerierSelect/SortedHead/10of1000000-4 4000089 4000089 +0.00% BenchmarkQuerierSelect/SortedHead/100of1000000-4 4000269 4000269 +0.00% BenchmarkQuerierSelect/SortedHead/1000of1000000-4 4002069 4002069 +0.00% BenchmarkQuerierSelect/SortedHead/10000of1000000-4 4020069 4020069 +0.00% BenchmarkQuerierSelect/SortedHead/100000of1000000-4 4200069 4200069 +0.00% BenchmarkQuerierSelect/SortedHead/1000000of1000000-4 6000069 6000069 +0.00% BenchmarkQuerierSelect/Block/1of1000000-4 6000023 6000022 -0.00% BenchmarkQuerierSelect/Block/10of1000000-4 6000059 6000058 -0.00% BenchmarkQuerierSelect/Block/100of1000000-4 6000419 6000418 -0.00% BenchmarkQuerierSelect/Block/1000of1000000-4 6004019 6004018 -0.00% BenchmarkQuerierSelect/Block/10000of1000000-4 6040019 6040018 -0.00% BenchmarkQuerierSelect/Block/100000of1000000-4 6400019 6400018 -0.00% BenchmarkQuerierSelect/Block/1000000of1000000-4 10000020 10000019 -0.00% benchmark old bytes new bytes delta BenchmarkQuerierSelect/Head/1of1000000-4 229192200 176001176 -23.21% BenchmarkQuerierSelect/Head/10of1000000-4 229193352 176002328 -23.21% BenchmarkQuerierSelect/Head/100of1000000-4 229204872 176013848 -23.21% BenchmarkQuerierSelect/Head/1000of1000000-4 229320072 176129048 -23.20% BenchmarkQuerierSelect/Head/10000of1000000-4 230472072 177281048 -23.08% BenchmarkQuerierSelect/Head/100000of1000000-4 241992072 188801048 -21.98% BenchmarkQuerierSelect/Head/1000000of1000000-4 357192072 304001048 -14.89% BenchmarkQuerierSelect/SortedHead/1of1000000-4 229193928 229193928 +0.00% BenchmarkQuerierSelect/SortedHead/10of1000000-4 229195080 229195080 +0.00% BenchmarkQuerierSelect/SortedHead/100of1000000-4 229206600 229206600 +0.00% BenchmarkQuerierSelect/SortedHead/1000of1000000-4 229321800 229321800 +0.00% BenchmarkQuerierSelect/SortedHead/10000of1000000-4 230473800 230473800 +0.00% BenchmarkQuerierSelect/SortedHead/100000of1000000-4 241993800 241993800 +0.00% BenchmarkQuerierSelect/SortedHead/1000000of1000000-4 357193800 357193800 +0.00% BenchmarkQuerierSelect/Block/1of1000000-4 227201516 227201500 -0.00% BenchmarkQuerierSelect/Block/10of1000000-4 227202924 227202908 -0.00% BenchmarkQuerierSelect/Block/100of1000000-4 227217036 227217020 -0.00% BenchmarkQuerierSelect/Block/1000of1000000-4 227358156 227358140 -0.00% BenchmarkQuerierSelect/Block/10000of1000000-4 228769356 228769340 -0.00% BenchmarkQuerierSelect/Block/100000of1000000-4 242881356 242881340 -0.00% BenchmarkQuerierSelect/Block/1000000of1000000-4 384001616 384001600 -0.00% Signed-off-by: Brian Brazil <brian.brazil@robustperception.io>
2020-01-17 03:21:44 -08:00
}
var (
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
seriesSets = make([]genericSeriesSet, 0, len(q.queriers))
wg sync.WaitGroup
seriesSetChan = make(chan genericSeriesSet)
)
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// Schedule all Selects for all queriers we know about.
2017-05-10 02:44:13 -07:00
for _, querier := range q.queriers {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
wg.Add(1)
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
go func(qr genericQuerier) {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
defer wg.Done()
// We need to sort for NewMergeSeriesSet to work.
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
seriesSetChan <- qr.Select(true, hints, matchers...)
}(querier)
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
go func() {
wg.Wait()
close(seriesSetChan)
}()
for r := range seriesSetChan {
seriesSets = append(seriesSets, r)
}
return &lazySeriesSet{create: create(seriesSets, q.mergeFn)}
}
func create(seriesSets []genericSeriesSet, mergeFunc genericSeriesMergeFunc) func() (genericSeriesSet, bool) {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// Returned function gets called with the first call to Next().
return func() (genericSeriesSet, bool) {
if len(seriesSets) == 1 {
return seriesSets[0], seriesSets[0].Next()
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
var h genericSeriesSetHeap
for _, set := range seriesSets {
if set == nil {
continue
}
if set.Next() {
heap.Push(&h, set)
continue
}
// When primary fails ignore results from secondaries.
// Only the primary querier returns error.
if err := set.Err(); err != nil {
return errorOnlySeriesSet{err}, false
}
2017-11-23 04:50:06 -08:00
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
set := &genericMergeSeriesSet{
mergeFunc: mergeFunc,
sets: seriesSets,
heap: h,
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
}
return set, set.Next()
2017-05-10 02:44:13 -07:00
}
}
// LabelValues returns all potential values for a label name.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (q *mergeGenericQuerier) LabelValues(name string) ([]string, Warnings, error) {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
var (
results [][]string
warnings Warnings
)
2017-05-10 02:44:13 -07:00
for _, querier := range q.queriers {
values, wrn, err := querier.LabelValues(name)
if wrn != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// TODO(bwplotka): We could potentially wrap warnings.
warnings = append(warnings, wrn...)
}
2017-05-10 02:44:13 -07:00
if err != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
return nil, nil, errors.Wrapf(err, "LabelValues() from Querier for label %s", name)
2017-05-10 02:44:13 -07:00
}
results = append(results, values)
}
return mergeStringSlices(results), warnings, nil
2017-05-10 02:44:13 -07:00
}
func mergeStringSlices(ss [][]string) []string {
switch len(ss) {
case 0:
return nil
case 1:
return ss[0]
case 2:
return mergeTwoStringSlices(ss[0], ss[1])
default:
halfway := len(ss) / 2
return mergeTwoStringSlices(
mergeStringSlices(ss[:halfway]),
mergeStringSlices(ss[halfway:]),
)
}
}
func mergeTwoStringSlices(a, b []string) []string {
i, j := 0, 0
result := make([]string, 0, len(a)+len(b))
for i < len(a) && j < len(b) {
switch strings.Compare(a[i], b[j]) {
case 0:
result = append(result, a[i])
i++
j++
case -1:
2017-05-10 02:44:13 -07:00
result = append(result, a[i])
i++
case 1:
2017-05-10 02:44:13 -07:00
result = append(result, b[j])
j++
}
}
result = append(result, a[i:]...)
result = append(result, b[j:]...)
2017-05-10 02:44:13 -07:00
return result
}
// LabelNames returns all the unique label names present in the block in sorted order.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (q *mergeGenericQuerier) LabelNames() ([]string, Warnings, error) {
var (
labelNamesMap = make(map[string]struct{})
warnings Warnings
)
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
for _, querier := range q.queriers {
names, wrn, err := querier.LabelNames()
if wrn != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// TODO(bwplotka): We could potentially wrap warnings.
warnings = append(warnings, wrn...)
}
if err != nil {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
return nil, nil, errors.Wrap(err, "LabelNames() from Querier")
}
for _, name := range names {
labelNamesMap[name] = struct{}{}
}
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
if len(labelNamesMap) == 0 {
return nil, warnings, nil
}
labelNames := make([]string, 0, len(labelNamesMap))
for name := range labelNamesMap {
labelNames = append(labelNames, name)
}
sort.Strings(labelNames)
return labelNames, warnings, nil
}
2017-05-10 02:44:13 -07:00
// Close releases the resources of the Querier.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (q *mergeGenericQuerier) Close() error {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
errs := tsdb_errors.MultiError{}
2017-05-10 02:44:13 -07:00
for _, querier := range q.queriers {
if err := querier.Close(); err != nil {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
errs.Add(err)
2017-05-10 02:44:13 -07:00
}
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return errs.Err()
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
// VerticalSeriesMergeFunc returns merged series implementation that merges series with same labels together.
// It has to handle time-overlapped series as well.
type VerticalSeriesMergeFunc func(...Series) Series
// NewMergeSeriesSet returns a new SeriesSet that merges many SeriesSets together.
func NewMergeSeriesSet(sets []SeriesSet, mergeFunc VerticalSeriesMergeFunc) SeriesSet {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
genericSets := make([]genericSeriesSet, 0, len(sets))
for _, s := range sets {
genericSets = append(genericSets, &genericSeriesSetAdapter{s})
}
return &seriesSetAdapter{newGenericMergeSeriesSet(genericSets, (&seriesMergerAdapter{VerticalSeriesMergeFunc: mergeFunc}).Merge)}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
// VerticalChunkSeriesMergeFunc returns merged chunk series implementation that merges potentially time-overlapping
// chunk series with the same labels into single ChunkSeries.
//
// NOTE: It's up to implementation how series are vertically merged (if chunks are sorted, re-encoded etc).
type VerticalChunkSeriesMergeFunc func(...ChunkSeries) ChunkSeries
// NewMergeChunkSeriesSet returns a new ChunkSeriesSet that merges many SeriesSet together.
func NewMergeChunkSeriesSet(sets []ChunkSeriesSet, mergeFunc VerticalChunkSeriesMergeFunc) ChunkSeriesSet {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
genericSets := make([]genericSeriesSet, 0, len(sets))
for _, s := range sets {
genericSets = append(genericSets, &genericChunkSeriesSetAdapter{s})
}
return &chunkSeriesSetAdapter{newGenericMergeSeriesSet(genericSets, (&chunkSeriesMergerAdapter{VerticalChunkSeriesMergeFunc: mergeFunc}).Merge)}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
}
// genericMergeSeriesSet implements genericSeriesSet.
type genericMergeSeriesSet struct {
currentLabels labels.Labels
mergeFunc genericSeriesMergeFunc
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
heap genericSeriesSetHeap
sets []genericSeriesSet
currentSets []genericSeriesSet
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
// newGenericMergeSeriesSet returns a new genericSeriesSet that merges (and deduplicates)
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// series returned by the series sets when iterating.
// Each series set must return its series in labels order, otherwise
// merged series set will be incorrect.
// Overlapped situations are merged using provided mergeFunc.
func newGenericMergeSeriesSet(sets []genericSeriesSet, mergeFunc genericSeriesMergeFunc) genericSeriesSet {
if len(sets) == 1 {
return sets[0]
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// We are pre-advancing sets, so we can introspect the label of the
2017-05-10 02:44:13 -07:00
// series under the cursor.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
var h genericSeriesSetHeap
2017-05-10 02:44:13 -07:00
for _, set := range sets {
if set == nil {
continue
}
2017-05-10 02:44:13 -07:00
if set.Next() {
heap.Push(&h, set)
}
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return &genericMergeSeriesSet{
mergeFunc: mergeFunc,
sets: sets,
heap: h,
2017-05-10 02:44:13 -07:00
}
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *genericMergeSeriesSet) Next() bool {
// Run in a loop because the "next" series sets may not be valid anymore.
// If, for the current label set, all the next series sets come from
// failed remote storage sources, we want to keep trying with the next label set.
for {
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
// Firstly advance all the current series sets. If any of them have run out
// we can drop them, otherwise they should be inserted back into the heap.
for _, set := range c.currentSets {
if set.Next() {
heap.Push(&c.heap, set)
}
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
if len(c.heap) == 0 {
return false
2017-05-10 02:44:13 -07:00
}
// Now, pop items of the heap that have equal label sets.
c.currentSets = nil
c.currentLabels = c.heap[0].At().Labels()
for len(c.heap) > 0 && labels.Equal(c.currentLabels, c.heap[0].At().Labels()) {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
set := heap.Pop(&c.heap).(genericSeriesSet)
c.currentSets = append(c.currentSets, set)
}
// As long as the current set contains at least 1 set,
// then it should return true.
if len(c.currentSets) != 0 {
break
}
2017-05-10 02:44:13 -07:00
}
return true
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *genericMergeSeriesSet) At() Labels {
if len(c.currentSets) == 1 {
return c.currentSets[0].At()
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
series := make([]Labels, 0, len(c.currentSets))
2017-05-10 02:44:13 -07:00
for _, seriesSet := range c.currentSets {
series = append(series, seriesSet.At())
}
return c.mergeFunc(series...)
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *genericMergeSeriesSet) Err() error {
2017-05-10 02:44:13 -07:00
for _, set := range c.sets {
if err := set.Err(); err != nil {
return err
}
}
return nil
}
*: Consistent Error/Warning handling for SeriesSet iterator: Allowing Async Select (#7251) * Add errors and Warnings to SeriesSet Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Change Querier interface and refactor accordingly Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor promql/engine to propagate warnings at eval stage Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Make sure all the series from all Selects are pre-advanced Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Separate merge series sets Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Clean Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactor merge querier failure handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Refactored and simplified fanout with improvements from incoming chunk iterator PRs. * Secondary logic is hidden, instead of weird failed series set logic we had. * Fanout is well commented * Fanout closing record all errors * MergeQuerier improved API (clearer) * deferredGenericMergeSeriesSet is not needed as we return no samples anyway for failed series sets (next = false). Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Fix formatting Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix CI issues Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Added final tests for error handling. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. * Moved hints in populate to be allocated only when needed. * Used sync.Once in secondary Querier to achieve all-or-nothing partial response logic. * Select after first Next is done will panic. NOTE: in lazySeriesSet in theory we could just panic, I think however we can totally just return error, it will panic in expand anyway. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Utilize errWithWarnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix recently introduced expansion issue Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add tests for secondary querier error handling Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Implement lazy merge Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Add name to test cases Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Reorganize Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Address review comments Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Remove redundant warnings Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> * Fix rebase mistake Signed-off-by: Kemal Akkoyun <kakkoyun@gmail.com> Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-06-09 09:57:31 -07:00
func (c *genericMergeSeriesSet) Warnings() Warnings {
var ws Warnings
for _, set := range c.sets {
ws = append(ws, set.Warnings()...)
}
return ws
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
type genericSeriesSetHeap []genericSeriesSet
2017-05-10 02:44:13 -07:00
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h genericSeriesSetHeap) Len() int { return len(h) }
func (h genericSeriesSetHeap) Swap(i, j int) { h[i], h[j] = h[j], h[i] }
2017-05-10 02:44:13 -07:00
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h genericSeriesSetHeap) Less(i, j int) bool {
2017-05-10 02:44:13 -07:00
a, b := h[i].At().Labels(), h[j].At().Labels()
return labels.Compare(a, b) < 0
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h *genericSeriesSetHeap) Push(x interface{}) {
*h = append(*h, x.(genericSeriesSet))
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h *genericSeriesSetHeap) Pop() interface{} {
2017-05-10 02:44:13 -07:00
old := *h
n := len(old)
x := old[n-1]
*h = old[0 : n-1]
return x
}
// ChainedSeriesMerge returns single series from many same, potentially overlapping series by chaining samples together.
// If one or more samples overlap, one sample from random overlapped ones is kept and all others with the same
// timestamp are dropped.
//
// This works the best with replicated series, where data from two series are exactly the same. This does not work well
// with "almost" the same data, e.g. from 2 Prometheus HA replicas. This is fine, since from the Prometheus perspective
// this never happens.
//
// NOTE: Use this only when you see potentially overlapping series, as this introduces small overhead to handle overlaps
// between series.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func ChainedSeriesMerge(s ...Series) Series {
if len(s) == 0 {
return nil
}
return &chainSeries{
labels: s[0].Labels(),
series: s,
}
}
type chainSeries struct {
2017-05-10 02:44:13 -07:00
labels labels.Labels
series []Series
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (m *chainSeries) Labels() labels.Labels {
2017-05-10 02:44:13 -07:00
return m.labels
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (m *chainSeries) Iterator() chunkenc.Iterator {
iterators := make([]chunkenc.Iterator, 0, len(m.series))
2017-05-10 02:44:13 -07:00
for _, s := range m.series {
iterators = append(iterators, s.Iterator())
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return newChainSampleIterator(iterators)
2017-05-10 02:44:13 -07:00
}
// chainSampleIterator is responsible to iterate over samples from different iterators of the same time series in timestamps
// order. If one or more samples overlap, one sample from random overlapped ones is kept and all others with the same
// timestamp are dropped.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
type chainSampleIterator struct {
iterators []chunkenc.Iterator
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
h samplesIteratorHeap
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func newChainSampleIterator(iterators []chunkenc.Iterator) chunkenc.Iterator {
return &chainSampleIterator{
2017-05-10 02:44:13 -07:00
iterators: iterators,
h: nil,
}
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *chainSampleIterator) Seek(t int64) bool {
c.h = samplesIteratorHeap{}
2017-05-10 02:44:13 -07:00
for _, iter := range c.iterators {
if iter.Seek(t) {
heap.Push(&c.h, iter)
}
}
return len(c.h) > 0
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *chainSampleIterator) At() (t int64, v float64) {
if len(c.h) == 0 {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
panic("chainSampleIterator.At() called after .Next() returned false.")
}
2017-05-10 02:44:13 -07:00
return c.h[0].At()
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *chainSampleIterator) Next() bool {
2017-05-10 02:44:13 -07:00
if c.h == nil {
for _, iter := range c.iterators {
if iter.Next() {
heap.Push(&c.h, iter)
}
}
return len(c.h) > 0
2017-05-10 02:44:13 -07:00
}
if len(c.h) == 0 {
return false
}
currt, _ := c.At()
for len(c.h) > 0 {
nextt, _ := c.h[0].At()
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
// All but one of the overlapping samples will be dropped.
if nextt != currt {
break
}
iter := heap.Pop(&c.h).(chunkenc.Iterator)
if iter.Next() {
heap.Push(&c.h, iter)
}
2017-05-10 02:44:13 -07:00
}
2017-05-10 02:44:13 -07:00
return len(c.h) > 0
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (c *chainSampleIterator) Err() error {
var errs tsdb_errors.MultiError
2017-05-10 02:44:13 -07:00
for _, iter := range c.iterators {
if err := iter.Err(); err != nil {
errs.Add(err)
2017-05-10 02:44:13 -07:00
}
}
return errs.Err()
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
type samplesIteratorHeap []chunkenc.Iterator
2017-05-10 02:44:13 -07:00
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h samplesIteratorHeap) Len() int { return len(h) }
func (h samplesIteratorHeap) Swap(i, j int) { h[i], h[j] = h[j], h[i] }
2017-05-10 02:44:13 -07:00
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h samplesIteratorHeap) Less(i, j int) bool {
2017-05-10 02:44:13 -07:00
at, _ := h[i].At()
bt, _ := h[j].At()
return at < bt
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h *samplesIteratorHeap) Push(x interface{}) {
*h = append(*h, x.(chunkenc.Iterator))
2017-05-10 02:44:13 -07:00
}
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
func (h *samplesIteratorHeap) Pop() interface{} {
old := *h
n := len(old)
x := old[n-1]
*h = old[0 : n-1]
return x
}
type compactChunkSeriesMerger struct {
mergeFunc VerticalSeriesMergeFunc
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
labels labels.Labels
series []ChunkSeries
}
// NewCompactingChunkSeriesMerger returns VerticalChunkSeriesMergeFunc that merges the same chunk series into single chunk series.
// In case of the chunk overlaps, it compacts those into one or more time-ordered non-overlapping chunks with merged data.
// Samples from overlapped chunks are merged using series vertical merge func.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
// It expects the same labels for each given series.
//
// NOTE: Use this only when you see potentially overlapping series, as this introduces small overhead to handle overlaps
// between series.
func NewCompactingChunkSeriesMerger(mergeFunc VerticalSeriesMergeFunc) VerticalChunkSeriesMergeFunc {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return func(s ...ChunkSeries) ChunkSeries {
if len(s) == 0 {
return nil
}
return &compactChunkSeriesMerger{
mergeFunc: mergeFunc,
labels: s[0].Labels(),
series: s,
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
}
}
func (s *compactChunkSeriesMerger) Labels() labels.Labels {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
return s.labels
}
func (s *compactChunkSeriesMerger) Iterator() chunks.Iterator {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
iterators := make([]chunks.Iterator, 0, len(s.series))
for _, series := range s.series {
iterators = append(iterators, series.Iterator())
}
return &compactChunkIterator{
mergeFunc: s.mergeFunc,
labels: s.labels,
iterators: iterators,
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
}
// compactChunkIterator is responsible to compact chunks from different iterators of the same time series into single chainSeries.
// If time-overlapping chunks are found, they are encoded and passed to series merge and encoded again into one bigger chunk.
// TODO(bwplotka): Currently merge will compact overlapping chunks with bigger chunk, without limit. Split it: https://github.com/prometheus/tsdb/issues/670
type compactChunkIterator struct {
mergeFunc VerticalSeriesMergeFunc
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
labels labels.Labels
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
iterators []chunks.Iterator
h chunkIteratorHeap
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
func (c *compactChunkIterator) At() chunks.Meta {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
if len(c.h) == 0 {
panic("compactChunkIterator.At() called after .Next() returned false.")
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
return c.h[0].At()
}
func (c *compactChunkIterator) Next() bool {
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
if c.h == nil {
for _, iter := range c.iterators {
if iter.Next() {
heap.Push(&c.h, iter)
}
}
return len(c.h) > 0
}
if len(c.h) == 0 {
return false
}
// Detect overlaps to compact.
// Be smart about it and deduplicate on the fly if chunks are identical.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
last := c.At()
var overlapped []Series
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
for {
iter := heap.Pop(&c.h).(chunks.Iterator)
if iter.Next() {
heap.Push(&c.h, iter)
}
if len(c.h) == 0 {
break
}
// Get the current oldest chunk by min, then max time.
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
next := c.At()
if next.MinTime > last.MaxTime {
// No overlap with last one.
break
}
if next.MinTime == last.MinTime &&
next.MaxTime == last.MaxTime &&
bytes.Equal(next.Chunk.Bytes(), last.Chunk.Bytes()) {
// 1:1 duplicates, skip last.
continue
}
overlapped = append(overlapped, &chunkToSeriesDecoder{
labels: c.labels,
Meta: last,
})
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
last = next
}
if len(overlapped) == 0 {
return len(c.h) > 0
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
// Add last, not yet included overlap.
overlapped = append(overlapped, &chunkToSeriesDecoder{
labels: c.labels,
Meta: c.At(),
})
var chkSeries ChunkSeries = &seriesToChunkEncoder{Series: c.mergeFunc(overlapped...)}
heap.Push(&c.h, chkSeries)
return true
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
func (c *compactChunkIterator) Err() error {
var errs tsdb_errors.MultiError
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
for _, iter := range c.iterators {
if err := iter.Err(); err != nil {
errs.Add(err)
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
}
return errs.Err()
storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. (#7005) * storage: Added Chunks{Queryable/Querier/SeriesSet/Series/Iteratable. Added generic Merge{SeriesSet/Querier} implementation. ## Rationales: In many places (e.g. chunk Remote read, Thanos Receive fetching chunk from TSDB), we operate on encoded chunks not samples. This means that we unnecessary decode/encode, wasting CPU, time and memory. This PR adds chunk iterator interfaces and makes the merge code to be reused between both seriesSets I will make the use of it in following PR inside tsdb itself. For now fanout implements it and mergers. All merges now also allows passing series mergers. This opens doors for custom deduplications other than TSDB vertical ones (e.g. offline one we have in Thanos). ## Changes * Added Chunk versions of all iterating methods. It all starts in Querier/ChunkQuerier. The plan is that Storage will implement both chunked and samples. * Added Seek to chunks.Iterator interface for iterating over chunks. * NewMergeChunkQuerier was added; Both this and NewMergeQuerier are now using generigMergeQuerier to share the code. Generic code was added. * Improved tests. * Added some TODO for further simplifications in next PRs. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Brian's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Moved s/Labeled/SeriesLabels as per Krasi suggestion. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Addressed Krasi's comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Second iteration of Krasi comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com> * Another round of comments. Signed-off-by: Bartlomiej Plotka <bwplotka@gmail.com>
2020-03-24 13:15:47 -07:00
}
type chunkIteratorHeap []chunks.Iterator
func (h chunkIteratorHeap) Len() int { return len(h) }
func (h chunkIteratorHeap) Swap(i, j int) { h[i], h[j] = h[j], h[i] }
func (h chunkIteratorHeap) Less(i, j int) bool {
at := h[i].At()
bt := h[j].At()
if at.MinTime == bt.MinTime {
return at.MaxTime < bt.MaxTime
}
return at.MinTime < bt.MinTime
}
func (h *chunkIteratorHeap) Push(x interface{}) {
*h = append(*h, x.(chunks.Iterator))
}
func (h *chunkIteratorHeap) Pop() interface{} {
2017-05-10 02:44:13 -07:00
old := *h
n := len(old)
x := old[n-1]
*h = old[0 : n-1]
return x
}