mirror of
https://github.com/prometheus/prometheus.git
synced 2024-12-28 06:59:40 -08:00
b3e34c6658
This commit introduces to Prometheus a batch database sample curator, which corroborates the high watermarks for sample series against the curation watermark table to see whether a curator of a given type needs to be run. The curator is an abstract executor, which runs various curation strategies across the database. It remarks the progress for each type of curation processor that runs for a given sample series. A curation procesor is responsible for effectuating the underlying batch changes that are request. In this commit, we introduce the CompactionProcessor, which takes several bits of runtime metadata and combine sparse sample entries in the database together to form larger groups. For instance, for a given series it would be possible to have the curator effectuate the following grouping: - Samples Older than Two Weeks: Grouped into Bunches of 10000 - Samples Older than One Week: Grouped into Bunches of 1000 - Samples Older than One Day: Grouped into Bunches of 100 - Samples Older than One Hour: Grouped into Bunches of 10 The benefits hereof of such a compaction are 1. a smaller search space in the database keyspace, 2. better employment of compression for repetious values, and 3. reduced seek times.
125 lines
4.2 KiB
Protocol Buffer
125 lines
4.2 KiB
Protocol Buffer
// Copyright 2013 Prometheus Team
|
|
// 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.
|
|
|
|
package io.prometheus;
|
|
|
|
import "google/protobuf/descriptor.proto";
|
|
|
|
message LabelPair {
|
|
optional string name = 1;
|
|
optional string value = 2;
|
|
}
|
|
|
|
message LabelName {
|
|
optional string name = 1;
|
|
}
|
|
|
|
message Metric {
|
|
repeated LabelPair label_pair = 1;
|
|
}
|
|
|
|
message Fingerprint {
|
|
optional string signature = 1;
|
|
}
|
|
|
|
message FingerprintCollection {
|
|
repeated Fingerprint member = 1;
|
|
}
|
|
|
|
message LabelSet {
|
|
repeated LabelPair member = 1;
|
|
}
|
|
|
|
// The default LevelDB comparator sorts not only lexicographically, but also by
|
|
// key length (which takes precedence). Thus, no variable-length fields may be
|
|
// introduced into the key definition below.
|
|
message SampleKey {
|
|
optional Fingerprint fingerprint = 1;
|
|
optional bytes timestamp = 2;
|
|
optional sfixed64 last_timestamp = 3;
|
|
optional fixed32 sample_count = 4;
|
|
}
|
|
|
|
message SampleValueSeries {
|
|
message Value {
|
|
optional int64 timestamp = 1;
|
|
optional double value = 2;
|
|
}
|
|
repeated Value value = 1;
|
|
}
|
|
|
|
message MembershipIndexValue {
|
|
}
|
|
|
|
message MetricHighWatermark {
|
|
optional int64 timestamp = 1;
|
|
}
|
|
|
|
// CompactionProcessorDefinition models a curation process across the sample
|
|
// corpus that ensures that sparse samples.
|
|
message CompactionProcessorDefinition {
|
|
// minimum_group_size identifies how minimally samples should be grouped
|
|
// together to write a new SampleValueSeries chunk.
|
|
optional uint32 minimum_group_size = 1;
|
|
}
|
|
|
|
// CurationKey models the state of curation for a given metric fingerprint and
|
|
// its associated samples. The time series database only knows about compaction
|
|
// and resampling behaviors that are explicitly defined to it in its runtime
|
|
// configuration, meaning it never scans on-disk tables for CurationKey
|
|
// policies; rather, it looks up via the CurationKey tuple to find out what the
|
|
// effectuation state for a given metric fingerprint is.
|
|
//
|
|
// For instance, how far along as a rule for (Fingerprint A, Samples Older Than
|
|
// B, and Curation Processor) has been effectuated on-disk.
|
|
message CurationKey {
|
|
// fingerprint identifies the fingerprint for the given policy.
|
|
optional Fingerprint fingerprint = 1;
|
|
|
|
// processor_message_type_name identifies the underlying message type that
|
|
// was used to encode processor_message_raw.
|
|
optional string processor_message_type_name = 2;
|
|
|
|
// processor_message_raw identifies the serialized ProcessorSignature for this
|
|
// operation.
|
|
optional bytes processor_message_raw = 3;
|
|
|
|
// ignore_younger_than represents in seconds relative to when the curation
|
|
// cycle start when the curator should stop operating. For instance, if
|
|
// the curation cycle starts at time T and the curation remark dictates that
|
|
// the curation should starts processing samples at time S, the curator should
|
|
// work from S until ignore_younger_than seconds before T:
|
|
//
|
|
// PAST NOW FUTURE
|
|
//
|
|
// S--------------->|----------T
|
|
// |---IYT----|
|
|
//
|
|
// [Curation Resumption Time (S), T - IYT)
|
|
optional int64 ignore_younger_than = 4;
|
|
|
|
// This could be populated by decoding the generated descriptor file into a
|
|
// FileDescriptorSet message and extracting the type definition for the given
|
|
// message schema that describes processor_message_type_name.
|
|
//
|
|
// optional google.protobuf.DescriptorProto processor_message_type_descriptor_raw = 5;
|
|
}
|
|
|
|
// CurationValue models the progress for a given CurationKey.
|
|
message CurationValue {
|
|
// last_completion_timestamp represents the seconds since the epoch UTC at
|
|
// which the curator last completed its duty cycle for a given metric
|
|
// fingerprint.
|
|
optional int64 last_completion_timestamp = 1;
|
|
}
|