2013-02-07 02:49:04 -08:00
|
|
|
// 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.
|
|
|
|
|
2013-01-07 14:24:26 -08:00
|
|
|
package ast
|
|
|
|
|
|
|
|
import (
|
2013-01-25 03:21:44 -08:00
|
|
|
"flag"
|
2013-06-25 05:02:27 -07:00
|
|
|
"time"
|
|
|
|
|
|
|
|
clientmodel "github.com/prometheus/client_golang/model"
|
|
|
|
|
2013-06-03 08:07:03 -07:00
|
|
|
"github.com/prometheus/prometheus/stats"
|
2013-01-27 09:49:45 -08:00
|
|
|
"github.com/prometheus/prometheus/storage/metric"
|
2013-01-07 14:24:26 -08:00
|
|
|
)
|
|
|
|
|
2013-01-25 03:21:44 -08:00
|
|
|
var defaultStalenessDelta = flag.Int("defaultStalenessDelta", 300, "Default staleness delta allowance in seconds during expression evaluations.")
|
2013-01-07 14:24:26 -08:00
|
|
|
|
2013-04-18 07:10:52 -07:00
|
|
|
// Describes the lenience limits to apply to values from the materialized view.
|
|
|
|
type StalenessPolicy struct {
|
|
|
|
// Describes the inclusive limit at which individual points if requested will
|
|
|
|
// be matched and subject to interpolation.
|
|
|
|
DeltaAllowance time.Duration
|
|
|
|
}
|
|
|
|
|
2013-03-21 10:06:15 -07:00
|
|
|
type viewAdapter struct {
|
2013-05-07 04:15:10 -07:00
|
|
|
// Policy that dictates when sample values around an evaluation time are to
|
|
|
|
// be interpreted as stale.
|
2013-04-18 07:10:52 -07:00
|
|
|
stalenessPolicy StalenessPolicy
|
2013-05-07 04:15:10 -07:00
|
|
|
// AST-global storage to use for operations that are not supported by views
|
|
|
|
// (i.e. fingerprint->metric lookups).
|
|
|
|
storage *metric.TieredStorage
|
|
|
|
// The materialized view which contains all timeseries data required for
|
|
|
|
// executing a query.
|
|
|
|
view metric.View
|
2013-06-03 08:07:03 -07:00
|
|
|
// The TimerGroup object in which to capture query timing statistics.
|
|
|
|
stats *stats.TimerGroup
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
|
|
|
|
2013-03-28 08:41:51 -07:00
|
|
|
// interpolateSamples interpolates a value at a target time between two
|
|
|
|
// provided sample pairs.
|
Use custom timestamp type for sample timestamps and related code.
So far we've been using Go's native time.Time for anything related to sample
timestamps. Since the range of time.Time is much bigger than what we need, this
has created two problems:
- there could be time.Time values which were out of the range/precision of the
time type that we persist to disk, therefore causing incorrectly ordered keys.
One bug caused by this was:
https://github.com/prometheus/prometheus/issues/367
It would be good to use a timestamp type that's more closely aligned with
what the underlying storage supports.
- sizeof(time.Time) is 192, while Prometheus should be ok with a single 64-bit
Unix timestamp (possibly even a 32-bit one). Since we store samples in large
numbers, this seriously affects memory usage. Furthermore, copying/working
with the data will be faster if it's smaller.
*MEMORY USAGE RESULTS*
Initial memory usage comparisons for a running Prometheus with 1 timeseries and
100,000 samples show roughly a 13% decrease in total (VIRT) memory usage. In my
tests, this advantage for some reason decreased a bit the more samples the
timeseries had (to 5-7% for millions of samples). This I can't fully explain,
but perhaps garbage collection issues were involved.
*WHEN TO USE THE NEW TIMESTAMP TYPE*
The new clientmodel.Timestamp type should be used whenever time
calculations are either directly or indirectly related to sample
timestamps.
For example:
- the timestamp of a sample itself
- all kinds of watermarks
- anything that may become or is compared to a sample timestamp (like the timestamp
passed into Target.Scrape()).
When to still use time.Time:
- for measuring durations/times not related to sample timestamps, like duration
telemetry exporting, timers that indicate how frequently to execute some
action, etc.
*NOTE ON OPERATOR OPTIMIZATION TESTS*
We don't use operator optimization code anymore, but it still lives in
the code as dead code. It still has tests, but I couldn't get all of them to
pass with the new timestamp format. I commented out the failing cases for now,
but we should probably remove the dead code soon. I just didn't want to do that
in the same change as this.
Change-Id: I821787414b0debe85c9fffaeb57abd453727af0f
2013-10-28 06:35:02 -07:00
|
|
|
func interpolateSamples(first, second *metric.SamplePair, timestamp clientmodel.Timestamp) *metric.SamplePair {
|
2013-03-28 08:41:51 -07:00
|
|
|
dv := second.Value - first.Value
|
|
|
|
dt := second.Timestamp.Sub(first.Timestamp)
|
|
|
|
|
2013-06-25 05:02:27 -07:00
|
|
|
dDt := dv / clientmodel.SampleValue(dt)
|
|
|
|
offset := clientmodel.SampleValue(timestamp.Sub(first.Timestamp))
|
2013-03-28 08:41:51 -07:00
|
|
|
|
2013-06-25 05:02:27 -07:00
|
|
|
return &metric.SamplePair{
|
2013-03-28 08:41:51 -07:00
|
|
|
Value: first.Value + (offset * dDt),
|
|
|
|
Timestamp: timestamp,
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// chooseClosestSample chooses the closest sample of a list of samples
|
|
|
|
// surrounding a given target time. If samples are found both before and after
|
|
|
|
// the target time, the sample value is interpolated between these. Otherwise,
|
|
|
|
// the single closest sample is returned verbatim.
|
Use custom timestamp type for sample timestamps and related code.
So far we've been using Go's native time.Time for anything related to sample
timestamps. Since the range of time.Time is much bigger than what we need, this
has created two problems:
- there could be time.Time values which were out of the range/precision of the
time type that we persist to disk, therefore causing incorrectly ordered keys.
One bug caused by this was:
https://github.com/prometheus/prometheus/issues/367
It would be good to use a timestamp type that's more closely aligned with
what the underlying storage supports.
- sizeof(time.Time) is 192, while Prometheus should be ok with a single 64-bit
Unix timestamp (possibly even a 32-bit one). Since we store samples in large
numbers, this seriously affects memory usage. Furthermore, copying/working
with the data will be faster if it's smaller.
*MEMORY USAGE RESULTS*
Initial memory usage comparisons for a running Prometheus with 1 timeseries and
100,000 samples show roughly a 13% decrease in total (VIRT) memory usage. In my
tests, this advantage for some reason decreased a bit the more samples the
timeseries had (to 5-7% for millions of samples). This I can't fully explain,
but perhaps garbage collection issues were involved.
*WHEN TO USE THE NEW TIMESTAMP TYPE*
The new clientmodel.Timestamp type should be used whenever time
calculations are either directly or indirectly related to sample
timestamps.
For example:
- the timestamp of a sample itself
- all kinds of watermarks
- anything that may become or is compared to a sample timestamp (like the timestamp
passed into Target.Scrape()).
When to still use time.Time:
- for measuring durations/times not related to sample timestamps, like duration
telemetry exporting, timers that indicate how frequently to execute some
action, etc.
*NOTE ON OPERATOR OPTIMIZATION TESTS*
We don't use operator optimization code anymore, but it still lives in
the code as dead code. It still has tests, but I couldn't get all of them to
pass with the new timestamp format. I commented out the failing cases for now,
but we should probably remove the dead code soon. I just didn't want to do that
in the same change as this.
Change-Id: I821787414b0debe85c9fffaeb57abd453727af0f
2013-10-28 06:35:02 -07:00
|
|
|
func (v *viewAdapter) chooseClosestSample(samples metric.Values, timestamp clientmodel.Timestamp) *metric.SamplePair {
|
2013-06-25 05:02:27 -07:00
|
|
|
var closestBefore *metric.SamplePair
|
|
|
|
var closestAfter *metric.SamplePair
|
2013-03-21 10:06:15 -07:00
|
|
|
for _, candidate := range samples {
|
2013-03-28 09:05:06 -07:00
|
|
|
delta := candidate.Timestamp.Sub(timestamp)
|
2013-03-28 08:41:51 -07:00
|
|
|
// Samples before target time.
|
2013-03-21 10:06:15 -07:00
|
|
|
if delta < 0 {
|
2013-03-28 08:41:51 -07:00
|
|
|
// Ignore samples outside of staleness policy window.
|
|
|
|
if -delta > v.stalenessPolicy.DeltaAllowance {
|
|
|
|
continue
|
|
|
|
}
|
|
|
|
// Ignore samples that are farther away than what we've seen before.
|
|
|
|
if closestBefore != nil && candidate.Timestamp.Before(closestBefore.Timestamp) {
|
|
|
|
continue
|
|
|
|
}
|
|
|
|
sample := candidate
|
2013-06-25 05:02:27 -07:00
|
|
|
closestBefore = sample
|
2013-03-21 10:06:15 -07:00
|
|
|
}
|
2013-01-07 14:24:26 -08:00
|
|
|
|
2013-03-28 08:41:51 -07:00
|
|
|
// Samples after target time.
|
|
|
|
if delta >= 0 {
|
|
|
|
// Ignore samples outside of staleness policy window.
|
|
|
|
if delta > v.stalenessPolicy.DeltaAllowance {
|
2013-03-21 10:06:15 -07:00
|
|
|
continue
|
|
|
|
}
|
2013-03-28 08:41:51 -07:00
|
|
|
// Ignore samples that are farther away than samples we've seen before.
|
|
|
|
if closestAfter != nil && candidate.Timestamp.After(closestAfter.Timestamp) {
|
|
|
|
continue
|
|
|
|
}
|
|
|
|
sample := candidate
|
2013-06-25 05:02:27 -07:00
|
|
|
closestAfter = sample
|
2013-03-21 10:06:15 -07:00
|
|
|
}
|
2013-03-28 08:41:51 -07:00
|
|
|
}
|
2013-03-21 10:06:15 -07:00
|
|
|
|
2013-03-28 08:41:51 -07:00
|
|
|
switch {
|
|
|
|
case closestBefore != nil && closestAfter != nil:
|
2013-05-15 22:38:31 -07:00
|
|
|
return interpolateSamples(closestBefore, closestAfter, timestamp)
|
2013-03-28 08:41:51 -07:00
|
|
|
case closestBefore != nil:
|
2013-05-15 22:38:31 -07:00
|
|
|
return closestBefore
|
2013-03-28 08:41:51 -07:00
|
|
|
default:
|
2013-05-15 22:38:31 -07:00
|
|
|
return closestAfter
|
2013-03-21 10:06:15 -07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Use custom timestamp type for sample timestamps and related code.
So far we've been using Go's native time.Time for anything related to sample
timestamps. Since the range of time.Time is much bigger than what we need, this
has created two problems:
- there could be time.Time values which were out of the range/precision of the
time type that we persist to disk, therefore causing incorrectly ordered keys.
One bug caused by this was:
https://github.com/prometheus/prometheus/issues/367
It would be good to use a timestamp type that's more closely aligned with
what the underlying storage supports.
- sizeof(time.Time) is 192, while Prometheus should be ok with a single 64-bit
Unix timestamp (possibly even a 32-bit one). Since we store samples in large
numbers, this seriously affects memory usage. Furthermore, copying/working
with the data will be faster if it's smaller.
*MEMORY USAGE RESULTS*
Initial memory usage comparisons for a running Prometheus with 1 timeseries and
100,000 samples show roughly a 13% decrease in total (VIRT) memory usage. In my
tests, this advantage for some reason decreased a bit the more samples the
timeseries had (to 5-7% for millions of samples). This I can't fully explain,
but perhaps garbage collection issues were involved.
*WHEN TO USE THE NEW TIMESTAMP TYPE*
The new clientmodel.Timestamp type should be used whenever time
calculations are either directly or indirectly related to sample
timestamps.
For example:
- the timestamp of a sample itself
- all kinds of watermarks
- anything that may become or is compared to a sample timestamp (like the timestamp
passed into Target.Scrape()).
When to still use time.Time:
- for measuring durations/times not related to sample timestamps, like duration
telemetry exporting, timers that indicate how frequently to execute some
action, etc.
*NOTE ON OPERATOR OPTIMIZATION TESTS*
We don't use operator optimization code anymore, but it still lives in
the code as dead code. It still has tests, but I couldn't get all of them to
pass with the new timestamp format. I commented out the failing cases for now,
but we should probably remove the dead code soon. I just didn't want to do that
in the same change as this.
Change-Id: I821787414b0debe85c9fffaeb57abd453727af0f
2013-10-28 06:35:02 -07:00
|
|
|
func (v *viewAdapter) GetValueAtTime(fingerprints clientmodel.Fingerprints, timestamp clientmodel.Timestamp) (Vector, error) {
|
2013-06-03 08:07:03 -07:00
|
|
|
timer := v.stats.GetTimer(stats.GetValueAtTimeTime).Start()
|
2013-07-25 03:16:32 -07:00
|
|
|
samples := Vector{}
|
2013-01-07 14:24:26 -08:00
|
|
|
for _, fingerprint := range fingerprints {
|
2013-03-28 09:05:06 -07:00
|
|
|
sampleCandidates := v.view.GetValueAtTime(fingerprint, timestamp)
|
2013-03-21 10:06:15 -07:00
|
|
|
samplePair := v.chooseClosestSample(sampleCandidates, timestamp)
|
2013-05-07 04:15:10 -07:00
|
|
|
m, err := v.storage.GetMetricForFingerprint(fingerprint)
|
2013-01-07 14:24:26 -08:00
|
|
|
if err != nil {
|
2013-07-25 03:16:32 -07:00
|
|
|
return nil, err
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
2013-03-21 10:06:15 -07:00
|
|
|
if samplePair != nil {
|
2013-06-25 05:02:27 -07:00
|
|
|
samples = append(samples, &clientmodel.Sample{
|
2013-05-14 07:25:06 -07:00
|
|
|
Metric: m,
|
2013-03-21 10:06:15 -07:00
|
|
|
Value: samplePair.Value,
|
2013-03-28 09:05:06 -07:00
|
|
|
Timestamp: timestamp,
|
2013-03-21 10:06:15 -07:00
|
|
|
})
|
|
|
|
}
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
2013-06-03 08:07:03 -07:00
|
|
|
timer.Stop()
|
2013-07-25 03:16:32 -07:00
|
|
|
return samples, nil
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
|
|
|
|
2013-07-25 03:16:32 -07:00
|
|
|
func (v *viewAdapter) GetBoundaryValues(fingerprints clientmodel.Fingerprints, interval *metric.Interval) ([]metric.SampleSet, error) {
|
2013-06-03 08:07:03 -07:00
|
|
|
timer := v.stats.GetTimer(stats.GetBoundaryValuesTime).Start()
|
2013-07-25 03:16:32 -07:00
|
|
|
sampleSets := []metric.SampleSet{}
|
2013-03-18 09:12:51 -07:00
|
|
|
for _, fingerprint := range fingerprints {
|
2013-05-11 17:03:16 -07:00
|
|
|
samplePairs := v.view.GetBoundaryValues(fingerprint, *interval)
|
2013-05-28 03:46:53 -07:00
|
|
|
if len(samplePairs) == 0 {
|
2013-03-21 10:06:15 -07:00
|
|
|
continue
|
2013-01-10 16:19:27 -08:00
|
|
|
}
|
2013-03-21 10:06:15 -07:00
|
|
|
|
|
|
|
// TODO: memoize/cache this.
|
2013-05-07 04:15:10 -07:00
|
|
|
m, err := v.storage.GetMetricForFingerprint(fingerprint)
|
2013-03-21 10:06:15 -07:00
|
|
|
if err != nil {
|
2013-07-25 03:16:32 -07:00
|
|
|
return nil, err
|
2013-01-10 16:19:27 -08:00
|
|
|
}
|
|
|
|
|
2013-06-25 05:02:27 -07:00
|
|
|
sampleSet := metric.SampleSet{
|
2013-05-14 07:25:06 -07:00
|
|
|
Metric: m,
|
2013-03-21 10:06:15 -07:00
|
|
|
Values: samplePairs,
|
|
|
|
}
|
2013-01-10 16:19:27 -08:00
|
|
|
sampleSets = append(sampleSets, sampleSet)
|
|
|
|
}
|
2013-06-03 08:07:03 -07:00
|
|
|
timer.Stop()
|
2013-01-10 16:19:27 -08:00
|
|
|
return sampleSets, nil
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
|
|
|
|
2013-07-25 03:16:32 -07:00
|
|
|
func (v *viewAdapter) GetRangeValues(fingerprints clientmodel.Fingerprints, interval *metric.Interval) ([]metric.SampleSet, error) {
|
2013-06-03 08:07:03 -07:00
|
|
|
timer := v.stats.GetTimer(stats.GetRangeValuesTime).Start()
|
2013-07-25 03:16:32 -07:00
|
|
|
sampleSets := []metric.SampleSet{}
|
2013-03-18 09:12:51 -07:00
|
|
|
for _, fingerprint := range fingerprints {
|
2013-03-21 10:06:15 -07:00
|
|
|
samplePairs := v.view.GetRangeValues(fingerprint, *interval)
|
2013-05-28 03:46:53 -07:00
|
|
|
if len(samplePairs) == 0 {
|
2013-03-21 10:06:15 -07:00
|
|
|
continue
|
2013-01-10 16:19:27 -08:00
|
|
|
}
|
2013-03-21 10:06:15 -07:00
|
|
|
|
|
|
|
// TODO: memoize/cache this.
|
2013-05-07 04:15:10 -07:00
|
|
|
m, err := v.storage.GetMetricForFingerprint(fingerprint)
|
2013-03-21 10:06:15 -07:00
|
|
|
if err != nil {
|
2013-07-25 03:16:32 -07:00
|
|
|
return nil, err
|
2013-01-10 16:19:27 -08:00
|
|
|
}
|
|
|
|
|
2013-06-25 05:02:27 -07:00
|
|
|
sampleSet := metric.SampleSet{
|
2013-05-14 07:25:06 -07:00
|
|
|
Metric: m,
|
2013-03-21 10:06:15 -07:00
|
|
|
Values: samplePairs,
|
|
|
|
}
|
2013-01-10 16:19:27 -08:00
|
|
|
sampleSets = append(sampleSets, sampleSet)
|
|
|
|
}
|
2013-06-03 08:07:03 -07:00
|
|
|
timer.Stop()
|
2013-01-10 16:19:27 -08:00
|
|
|
return sampleSets, nil
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
|
|
|
|
2013-06-03 08:07:03 -07:00
|
|
|
func NewViewAdapter(view metric.View, storage *metric.TieredStorage, queryStats *stats.TimerGroup) *viewAdapter {
|
2013-04-18 07:10:52 -07:00
|
|
|
stalenessPolicy := StalenessPolicy{
|
2013-03-21 10:06:15 -07:00
|
|
|
DeltaAllowance: time.Duration(*defaultStalenessDelta) * time.Second,
|
2013-01-25 03:21:44 -08:00
|
|
|
}
|
2013-03-21 10:06:15 -07:00
|
|
|
|
|
|
|
return &viewAdapter{
|
2013-04-18 07:10:52 -07:00
|
|
|
stalenessPolicy: stalenessPolicy,
|
2013-05-07 04:15:10 -07:00
|
|
|
storage: storage,
|
|
|
|
view: view,
|
2013-06-03 08:07:03 -07:00
|
|
|
stats: queryStats,
|
2013-01-07 14:24:26 -08:00
|
|
|
}
|
|
|
|
}
|