2013-02-08 17:03:26 +00:00
|
|
|
// 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.
|
|
|
|
|
|
|
|
package metric
|
|
|
|
|
|
|
|
import (
|
|
|
|
"sort"
|
|
|
|
"time"
|
2013-06-19 13:00:00 +00:00
|
|
|
|
2013-06-25 12:02:27 +00:00
|
|
|
clientmodel "github.com/prometheus/client_golang/model"
|
2013-02-08 17:03:26 +00:00
|
|
|
)
|
|
|
|
|
|
|
|
var (
|
|
|
|
// firstSupertime is the smallest valid supertime that may be seeked to.
|
|
|
|
firstSupertime = []byte{0, 0, 0, 0, 0, 0, 0, 0}
|
|
|
|
// lastSupertime is the largest valid supertime that may be seeked to.
|
|
|
|
lastSupertime = []byte{127, 255, 255, 255, 255, 255, 255, 255}
|
|
|
|
)
|
|
|
|
|
|
|
|
// Represents the summation of all datastore queries that shall be performed to
|
|
|
|
// extract values. Each operation mutates the state of the builder.
|
|
|
|
type ViewRequestBuilder interface {
|
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 13:35:02 +00:00
|
|
|
GetMetricAtTime(fingerprint *clientmodel.Fingerprint, time clientmodel.Timestamp)
|
|
|
|
GetMetricAtInterval(fingerprint *clientmodel.Fingerprint, from, through clientmodel.Timestamp, interval time.Duration)
|
|
|
|
GetMetricRange(fingerprint *clientmodel.Fingerprint, from, through clientmodel.Timestamp)
|
2013-02-08 17:03:26 +00:00
|
|
|
ScanJobs() scanJobs
|
|
|
|
}
|
|
|
|
|
|
|
|
// Contains the various unoptimized requests for data.
|
|
|
|
type viewRequestBuilder struct {
|
2013-06-25 12:02:27 +00:00
|
|
|
operations map[clientmodel.Fingerprint]ops
|
2013-02-08 17:03:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Furnishes a ViewRequestBuilder for remarking what types of queries to perform.
|
2013-05-07 12:25:01 +00:00
|
|
|
func NewViewRequestBuilder() *viewRequestBuilder {
|
|
|
|
return &viewRequestBuilder{
|
2013-06-25 12:02:27 +00:00
|
|
|
operations: make(map[clientmodel.Fingerprint]ops),
|
2013-02-08 17:03:26 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-08-29 13:15:22 +00:00
|
|
|
var getValuesAtTimes = newValueAtTimeList(10 * 1024)
|
|
|
|
|
2013-02-08 17:03:26 +00:00
|
|
|
// Gets for the given Fingerprint either the value at that time if there is an
|
|
|
|
// match or the one or two values adjacent thereto.
|
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 13:35:02 +00:00
|
|
|
func (v *viewRequestBuilder) GetMetricAtTime(fingerprint *clientmodel.Fingerprint, time clientmodel.Timestamp) {
|
2013-06-25 12:02:27 +00:00
|
|
|
ops := v.operations[*fingerprint]
|
2013-08-29 13:15:22 +00:00
|
|
|
op, _ := getValuesAtTimes.Get()
|
|
|
|
op.time = time
|
|
|
|
ops = append(ops, op)
|
2013-06-25 12:02:27 +00:00
|
|
|
v.operations[*fingerprint] = ops
|
2013-02-08 17:03:26 +00:00
|
|
|
}
|
|
|
|
|
2013-08-29 13:15:22 +00:00
|
|
|
var getValuesAtIntervals = newValueAtIntervalList(10 * 1024)
|
|
|
|
|
2013-02-08 17:03:26 +00:00
|
|
|
// Gets for the given Fingerprint either the value at that interval from From
|
|
|
|
// through Through if there is an match or the one or two values adjacent
|
|
|
|
// for each point.
|
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 13:35:02 +00:00
|
|
|
func (v *viewRequestBuilder) GetMetricAtInterval(fingerprint *clientmodel.Fingerprint, from, through clientmodel.Timestamp, interval time.Duration) {
|
2013-06-25 12:02:27 +00:00
|
|
|
ops := v.operations[*fingerprint]
|
2013-08-29 13:15:22 +00:00
|
|
|
op, _ := getValuesAtIntervals.Get()
|
|
|
|
op.from = from
|
|
|
|
op.through = through
|
|
|
|
op.interval = interval
|
|
|
|
ops = append(ops, op)
|
2013-06-25 12:02:27 +00:00
|
|
|
v.operations[*fingerprint] = ops
|
2013-02-08 17:03:26 +00:00
|
|
|
}
|
|
|
|
|
2013-08-29 13:15:22 +00:00
|
|
|
var getValuesAlongRanges = newValueAlongRangeList(10 * 1024)
|
|
|
|
|
2013-05-07 12:25:01 +00:00
|
|
|
// Gets for the given Fingerprint the values that occur inclusively from From
|
|
|
|
// through Through.
|
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 13:35:02 +00:00
|
|
|
func (v *viewRequestBuilder) GetMetricRange(fingerprint *clientmodel.Fingerprint, from, through clientmodel.Timestamp) {
|
2013-06-25 12:02:27 +00:00
|
|
|
ops := v.operations[*fingerprint]
|
2013-08-29 13:15:22 +00:00
|
|
|
op, _ := getValuesAlongRanges.Get()
|
|
|
|
op.from = from
|
|
|
|
op.through = through
|
|
|
|
ops = append(ops, op)
|
2013-06-25 12:02:27 +00:00
|
|
|
v.operations[*fingerprint] = ops
|
2013-02-08 17:03:26 +00:00
|
|
|
}
|
|
|
|
|
2013-08-29 13:15:22 +00:00
|
|
|
var getValuesAtIntervalAlongRanges = newValueAtIntervalAlongRangeList(10 * 1024)
|
|
|
|
|
2013-05-07 12:25:01 +00:00
|
|
|
// Gets value ranges at intervals for the given Fingerprint:
|
|
|
|
//
|
|
|
|
// |----| |----| |----| |----|
|
|
|
|
// ^ ^ ^ ^ ^ ^
|
|
|
|
// | \------------/ \----/ |
|
|
|
|
// from interval rangeDuration through
|
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 13:35:02 +00:00
|
|
|
func (v *viewRequestBuilder) GetMetricRangeAtInterval(fingerprint *clientmodel.Fingerprint, from, through clientmodel.Timestamp, interval, rangeDuration time.Duration) {
|
2013-05-07 12:25:01 +00:00
|
|
|
ops := v.operations[*fingerprint]
|
2013-08-29 13:15:22 +00:00
|
|
|
op, _ := getValuesAtIntervalAlongRanges.Get()
|
|
|
|
op.rangeFrom = from
|
|
|
|
op.rangeThrough = from.Add(rangeDuration)
|
|
|
|
op.rangeDuration = rangeDuration
|
|
|
|
op.interval = interval
|
|
|
|
op.through = through
|
|
|
|
ops = append(ops, op)
|
2013-05-07 12:25:01 +00:00
|
|
|
v.operations[*fingerprint] = ops
|
|
|
|
}
|
|
|
|
|
2013-02-08 17:03:26 +00:00
|
|
|
// Emits the optimized scans that will occur in the data store. This
|
|
|
|
// effectively resets the ViewRequestBuilder back to a pristine state.
|
2013-05-07 12:25:01 +00:00
|
|
|
func (v *viewRequestBuilder) ScanJobs() (j scanJobs) {
|
2013-02-08 17:03:26 +00:00
|
|
|
for fingerprint, operations := range v.operations {
|
2013-03-15 20:05:51 +00:00
|
|
|
sort.Sort(startsAtSort{operations})
|
2013-02-08 17:03:26 +00:00
|
|
|
|
2013-05-17 10:58:15 +00:00
|
|
|
fpCopy := fingerprint
|
2013-02-08 17:03:26 +00:00
|
|
|
j = append(j, scanJob{
|
2013-05-17 10:58:15 +00:00
|
|
|
fingerprint: &fpCopy,
|
2013-05-07 12:25:01 +00:00
|
|
|
// BUG: Evaluate whether we need to implement an optimize() working also
|
|
|
|
// for getValueRangeAtIntervalOp and use it here instead of just passing
|
|
|
|
// through the list of ops as-is.
|
|
|
|
operations: operations,
|
2013-02-08 17:03:26 +00:00
|
|
|
})
|
|
|
|
|
|
|
|
delete(v.operations, fingerprint)
|
|
|
|
}
|
|
|
|
|
|
|
|
sort.Sort(j)
|
|
|
|
|
|
|
|
return
|
|
|
|
}
|
2013-03-07 01:16:39 +00:00
|
|
|
|
|
|
|
type view struct {
|
2013-05-16 14:02:07 +00:00
|
|
|
*memorySeriesStorage
|
2013-03-07 01:16:39 +00:00
|
|
|
}
|
|
|
|
|
2013-06-25 12:02:27 +00:00
|
|
|
func (v view) appendSamples(fingerprint *clientmodel.Fingerprint, samples Values) {
|
2013-05-28 12:36:03 +00:00
|
|
|
v.memorySeriesStorage.appendSamplesWithoutIndexing(fingerprint, samples)
|
2013-03-07 01:16:39 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
func newView() view {
|
2013-06-06 16:16:22 +00:00
|
|
|
return view{NewMemorySeriesStorage(MemorySeriesOptions{})}
|
2013-03-07 01:16:39 +00:00
|
|
|
}
|
2013-08-29 13:15:22 +00:00
|
|
|
|
|
|
|
func giveBackOp(op interface{}) bool {
|
|
|
|
switch v := op.(type) {
|
|
|
|
case *getValuesAtTimeOp:
|
|
|
|
return getValuesAtTimes.Give(v)
|
|
|
|
case *getValuesAtIntervalOp:
|
|
|
|
return getValuesAtIntervals.Give(v)
|
|
|
|
case *getValuesAlongRangeOp:
|
|
|
|
return getValuesAlongRanges.Give(v)
|
|
|
|
case *getValueRangeAtIntervalOp:
|
|
|
|
return getValuesAtIntervalAlongRanges.Give(v)
|
|
|
|
default:
|
|
|
|
panic("unrecognized operation")
|
|
|
|
}
|
|
|
|
}
|