client_golang/prometheus/vec.go

473 lines
13 KiB
Go
Raw Normal View History

// Copyright 2014 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.
package prometheus
import (
"fmt"
"sync"
"github.com/prometheus/common/model"
)
// metricVec is a Collector to bundle metrics of the same name that differ in
// their label values. metricVec is not used directly (and therefore
// unexported). It is used as a building block for implementations of vectors of
// a given metric type, like GaugeVec, CounterVec, SummaryVec, and HistogramVec.
// It also handles label currying. It uses basicMetricVec internally.
type metricVec struct {
*metricMap
curry []curriedLabelValue
// hashAdd and hashAddByte can be replaced for testing collision handling.
hashAdd func(h uint64, s string) uint64
hashAddByte func(h uint64, b byte) uint64
}
// newMetricVec returns an initialized metricVec.
func newMetricVec(desc *Desc, newMetric func(lvs ...string) Metric) *metricVec {
return &metricVec{
metricMap: &metricMap{
metrics: map[uint64][]metricWithLabelValues{},
desc: desc,
newMetric: newMetric,
},
hashAdd: hashAdd,
hashAddByte: hashAddByte,
}
}
// DeleteLabelValues removes the metric where the variable labels are the same
// as those passed in as labels (same order as the VariableLabels in Desc). It
// returns true if a metric was deleted.
//
// It is not an error if the number of label values is not the same as the
// number of VariableLabels in Desc. However, such inconsistent label count can
// never match an actual metric, so the method will always return false in that
// case.
//
// Note that for more than one label value, this method is prone to mistakes
// caused by an incorrect order of arguments. Consider Delete(Labels) as an
// alternative to avoid that type of mistake. For higher label numbers, the
// latter has a much more readable (albeit more verbose) syntax, but it comes
// with a performance overhead (for creating and processing the Labels map).
// See also the CounterVec example.
func (m *metricVec) DeleteLabelValues(lvs ...string) bool {
h, err := m.hashLabelValues(lvs)
if err != nil {
return false
}
return m.metricMap.deleteByHashWithLabelValues(h, lvs, m.curry)
}
// Delete deletes the metric where the variable labels are the same as those
// passed in as labels. It returns true if a metric was deleted.
//
// It is not an error if the number and names of the Labels are inconsistent
// with those of the VariableLabels in Desc. However, such inconsistent Labels
// can never match an actual metric, so the method will always return false in
// that case.
//
// This method is used for the same purpose as DeleteLabelValues(...string). See
// there for pros and cons of the two methods.
func (m *metricVec) Delete(labels Labels) bool {
h, err := m.hashLabels(labels)
if err != nil {
return false
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
return m.metricMap.deleteByHashWithLabels(h, labels, m.curry)
}
func (m *metricVec) curryWith(labels Labels) (*metricVec, error) {
var (
newCurry []curriedLabelValue
oldCurry = m.curry
iCurry int
)
for i, label := range m.desc.variableLabels {
val, ok := labels[label]
if iCurry < len(oldCurry) && oldCurry[iCurry].index == i {
if ok {
return nil, fmt.Errorf("label name %q is already curried", label)
}
newCurry = append(newCurry, oldCurry[iCurry])
iCurry++
} else {
if !ok {
continue // Label stays uncurried.
}
newCurry = append(newCurry, curriedLabelValue{i, val})
}
}
if l := len(oldCurry) + len(labels) - len(newCurry); l > 0 {
return nil, fmt.Errorf("%d unknown label(s) found during currying", l)
}
return &metricVec{
metricMap: m.metricMap,
curry: newCurry,
hashAdd: m.hashAdd,
hashAddByte: m.hashAddByte,
}, nil
}
func (m *metricVec) getMetricWithLabelValues(lvs ...string) (Metric, error) {
h, err := m.hashLabelValues(lvs)
if err != nil {
return nil, err
}
return m.metricMap.getOrCreateMetricWithLabelValues(h, lvs, m.curry), nil
}
func (m *metricVec) getMetricWith(labels Labels) (Metric, error) {
h, err := m.hashLabels(labels)
if err != nil {
return nil, err
}
return m.metricMap.getOrCreateMetricWithLabels(h, labels, m.curry), nil
}
func (m *metricVec) hashLabelValues(vals []string) (uint64, error) {
if err := validateLabelValues(vals, len(m.desc.variableLabels)-len(m.curry)); err != nil {
return 0, err
}
var (
h = hashNew()
curry = m.curry
iVals, iCurry int
)
for i := 0; i < len(m.desc.variableLabels); i++ {
if iCurry < len(curry) && curry[iCurry].index == i {
h = m.hashAdd(h, curry[iCurry].value)
iCurry++
} else {
h = m.hashAdd(h, vals[iVals])
iVals++
}
h = m.hashAddByte(h, model.SeparatorByte)
}
return h, nil
}
func (m *metricVec) hashLabels(labels Labels) (uint64, error) {
if err := validateValuesInLabels(labels, len(m.desc.variableLabels)-len(m.curry)); err != nil {
return 0, err
}
var (
h = hashNew()
curry = m.curry
iCurry int
)
for i, label := range m.desc.variableLabels {
val, ok := labels[label]
if iCurry < len(curry) && curry[iCurry].index == i {
if ok {
return 0, fmt.Errorf("label name %q is already curried", label)
}
h = m.hashAdd(h, curry[iCurry].value)
iCurry++
} else {
if !ok {
return 0, fmt.Errorf("label name %q missing in label map", label)
}
h = m.hashAdd(h, val)
}
h = m.hashAddByte(h, model.SeparatorByte)
}
return h, nil
}
// metricWithLabelValues provides the metric and its label values for
// disambiguation on hash collision.
type metricWithLabelValues struct {
values []string
metric Metric
}
// curriedLabelValue sets the curried value for a label at the given index.
type curriedLabelValue struct {
index int
value string
}
// metricMap is a helper for metricVec and shared between differently curried
// metricVecs.
type metricMap struct {
mtx sync.RWMutex // Protects metrics.
metrics map[uint64][]metricWithLabelValues
desc *Desc
newMetric func(labelValues ...string) Metric
}
// Describe implements Collector. It will send exactly one Desc to the provided
// channel.
func (m *metricMap) Describe(ch chan<- *Desc) {
ch <- m.desc
}
// Collect implements Collector.
func (m *metricMap) Collect(ch chan<- Metric) {
m.mtx.RLock()
defer m.mtx.RUnlock()
for _, metrics := range m.metrics {
for _, metric := range metrics {
ch <- metric.metric
}
}
}
// Clear deletes all metrics in this vector.
func (m *metricMap) Clear() {
m.mtx.Lock()
defer m.mtx.Unlock()
for h := range m.metrics {
delete(m.metrics, h)
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
// deleteByHashWithLabelValues removes the metric from the hash bucket h. If
// there are multiple matches in the bucket, use lvs to select a metric and
// remove only that metric.
func (m *metricMap) deleteByHashWithLabelValues(
h uint64, lvs []string, curry []curriedLabelValue,
) bool {
m.mtx.Lock()
defer m.mtx.Unlock()
metrics, ok := m.metrics[h]
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
if !ok {
return false
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
i := findMetricWithLabelValues(metrics, lvs, curry)
if i >= len(metrics) {
return false
}
if len(metrics) > 1 {
m.metrics[h] = append(metrics[:i], metrics[i+1:]...)
} else {
delete(m.metrics, h)
}
return true
}
// deleteByHashWithLabels removes the metric from the hash bucket h. If there
// are multiple matches in the bucket, use lvs to select a metric and remove
// only that metric.
func (m *metricMap) deleteByHashWithLabels(
h uint64, labels Labels, curry []curriedLabelValue,
) bool {
2018-04-09 17:07:30 +03:00
m.mtx.Lock()
defer m.mtx.Unlock()
metrics, ok := m.metrics[h]
if !ok {
return false
}
i := findMetricWithLabels(m.desc, metrics, labels, curry)
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
if i >= len(metrics) {
return false
}
if len(metrics) > 1 {
m.metrics[h] = append(metrics[:i], metrics[i+1:]...)
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
} else {
delete(m.metrics, h)
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
return true
}
// getOrCreateMetricWithLabelValues retrieves the metric by hash and label value
// or creates it and returns the new one.
//
// This function holds the mutex.
func (m *metricMap) getOrCreateMetricWithLabelValues(
hash uint64, lvs []string, curry []curriedLabelValue,
) Metric {
m.mtx.RLock()
metric, ok := m.getMetricWithHashAndLabelValues(hash, lvs, curry)
m.mtx.RUnlock()
if ok {
return metric
}
m.mtx.Lock()
defer m.mtx.Unlock()
metric, ok = m.getMetricWithHashAndLabelValues(hash, lvs, curry)
if !ok {
inlinedLVs := inlineLabelValues(lvs, curry)
metric = m.newMetric(inlinedLVs...)
m.metrics[hash] = append(m.metrics[hash], metricWithLabelValues{values: inlinedLVs, metric: metric})
}
return metric
}
// getOrCreateMetricWithLabelValues retrieves the metric by hash and label value
// or creates it and returns the new one.
//
// This function holds the mutex.
func (m *metricMap) getOrCreateMetricWithLabels(
hash uint64, labels Labels, curry []curriedLabelValue,
) Metric {
m.mtx.RLock()
metric, ok := m.getMetricWithHashAndLabels(hash, labels, curry)
m.mtx.RUnlock()
if ok {
return metric
}
m.mtx.Lock()
defer m.mtx.Unlock()
metric, ok = m.getMetricWithHashAndLabels(hash, labels, curry)
if !ok {
lvs := extractLabelValues(m.desc, labels, curry)
metric = m.newMetric(lvs...)
m.metrics[hash] = append(m.metrics[hash], metricWithLabelValues{values: lvs, metric: metric})
}
return metric
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
// getMetricWithHashAndLabelValues gets a metric while handling possible
// collisions in the hash space. Must be called while holding the read mutex.
func (m *metricMap) getMetricWithHashAndLabelValues(
h uint64, lvs []string, curry []curriedLabelValue,
) (Metric, bool) {
metrics, ok := m.metrics[h]
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
if ok {
if i := findMetricWithLabelValues(metrics, lvs, curry); i < len(metrics) {
return metrics[i].metric, true
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
return nil, false
}
// getMetricWithHashAndLabels gets a metric while handling possible collisions in
// the hash space. Must be called while holding read mutex.
func (m *metricMap) getMetricWithHashAndLabels(
h uint64, labels Labels, curry []curriedLabelValue,
) (Metric, bool) {
metrics, ok := m.metrics[h]
if ok {
if i := findMetricWithLabels(m.desc, metrics, labels, curry); i < len(metrics) {
return metrics[i].metric, true
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
return nil, false
}
// findMetricWithLabelValues returns the index of the matching metric or
// len(metrics) if not found.
func findMetricWithLabelValues(
metrics []metricWithLabelValues, lvs []string, curry []curriedLabelValue,
) int {
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
for i, metric := range metrics {
if matchLabelValues(metric.values, lvs, curry) {
return i
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
}
return len(metrics)
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
// findMetricWithLabels returns the index of the matching metric or len(metrics)
// if not found.
func findMetricWithLabels(
desc *Desc, metrics []metricWithLabelValues, labels Labels, curry []curriedLabelValue,
) int {
for i, metric := range metrics {
if matchLabels(desc, metric.values, labels, curry) {
return i
}
}
return len(metrics)
}
func matchLabelValues(values []string, lvs []string, curry []curriedLabelValue) bool {
if len(values) != len(lvs)+len(curry) {
return false
}
var iLVs, iCurry int
for i, v := range values {
if iCurry < len(curry) && curry[iCurry].index == i {
if v != curry[iCurry].value {
return false
}
iCurry++
continue
}
if v != lvs[iLVs] {
return false
}
iLVs++
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
return true
}
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
func matchLabels(desc *Desc, values []string, labels Labels, curry []curriedLabelValue) bool {
if len(values) != len(labels)+len(curry) {
return false
}
iCurry := 0
for i, k := range desc.variableLabels {
if iCurry < len(curry) && curry[iCurry].index == i {
if values[i] != curry[iCurry].value {
return false
}
iCurry++
continue
}
if values[i] != labels[k] {
return false
}
}
return true
}
func extractLabelValues(desc *Desc, labels Labels, curry []curriedLabelValue) []string {
labelValues := make([]string, len(labels)+len(curry))
iCurry := 0
for i, k := range desc.variableLabels {
if iCurry < len(curry) && curry[iCurry].index == i {
labelValues[i] = curry[iCurry].value
iCurry++
continue
}
labelValues[i] = labels[k]
}
return labelValues
metricvec: handle hash collision for labeled metrics While hash collisions are quite rare, the current state of the client library carries a risk of merging two separate label values into a single metric bucket. The effects are near impossible to detect and the result will be missing or incorrect counters. This changeset handles hash collisions by falling back to collision resolution if multiple label values hash to the same value. This works similar to separate chaining using a slice. Extra storage is minimized to only the value key slice to that metrics can be differentiated within a bucket. In general, the cost of handling collisions is completely minimized under normal operation. Performance does show slight increases in certain areas, but these are more likely statistically anomalies. More importantly, zero allocation behavior for metrics is preserved on the fast path. Minimal allocations may be made during collision handling but this has minimal effect. Benchmark comparisons with and without collision resolution follow. ``` benchmark old ns/op new ns/op delta BenchmarkCounterWithLabelValues-4 99.0 107 +8.08% BenchmarkCounterWithLabelValuesConcurrent-4 79.6 91.0 +14.32% BenchmarkCounterWithMappedLabels-4 518 542 +4.63% BenchmarkCounterWithPreparedMappedLabels-4 127 137 +7.87% BenchmarkCounterNoLabels-4 19.5 19.1 -2.05% BenchmarkGaugeWithLabelValues-4 97.4 110 +12.94% BenchmarkGaugeNoLabels-4 12.4 10.3 -16.94% BenchmarkSummaryWithLabelValues-4 1204 915 -24.00% BenchmarkSummaryNoLabels-4 936 847 -9.51% BenchmarkHistogramWithLabelValues-4 147 147 +0.00% BenchmarkHistogramNoLabels-4 50.6 49.3 -2.57% BenchmarkHistogramObserve1-4 37.9 37.5 -1.06% BenchmarkHistogramObserve2-4 122 137 +12.30% BenchmarkHistogramObserve4-4 310 352 +13.55% BenchmarkHistogramObserve8-4 691 729 +5.50% BenchmarkHistogramWrite1-4 3374 3097 -8.21% BenchmarkHistogramWrite2-4 5310 5051 -4.88% BenchmarkHistogramWrite4-4 12094 10690 -11.61% BenchmarkHistogramWrite8-4 19416 17755 -8.55% BenchmarkHandler-4 11934304 13765894 +15.35% BenchmarkSummaryObserve1-4 1119 1105 -1.25% BenchmarkSummaryObserve2-4 3679 3430 -6.77% BenchmarkSummaryObserve4-4 10678 7982 -25.25% BenchmarkSummaryObserve8-4 22974 16689 -27.36% BenchmarkSummaryWrite1-4 25962 14680 -43.46% BenchmarkSummaryWrite2-4 38019 35073 -7.75% BenchmarkSummaryWrite4-4 78027 56816 -27.18% BenchmarkSummaryWrite8-4 117220 132248 +12.82% BenchmarkMetricVecWithLabelValuesBasic-4 138 133 -3.62% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 150 144 -4.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 263 256 -2.66% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 145 155 +6.90% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 606 634 +4.62% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 746 641 -14.08% benchmark old allocs new allocs delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 2 2 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% benchmark old bytes new bytes delta BenchmarkCounterWithLabelValues-4 0 0 +0.00% BenchmarkCounterWithLabelValuesConcurrent-4 0 0 +0.00% BenchmarkCounterWithMappedLabels-4 336 336 +0.00% BenchmarkCounterWithPreparedMappedLabels-4 0 0 +0.00% BenchmarkCounterNoLabels-4 0 0 +0.00% BenchmarkGaugeWithLabelValues-4 0 0 +0.00% BenchmarkGaugeNoLabels-4 0 0 +0.00% BenchmarkSummaryWithLabelValues-4 0 0 +0.00% BenchmarkSummaryNoLabels-4 0 0 +0.00% BenchmarkHistogramWithLabelValues-4 0 0 +0.00% BenchmarkHistogramNoLabels-4 0 0 +0.00% BenchmarkMetricVecWithLabelValuesBasic-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues4Keys10ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues2Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys100ValueCardinality-4 0 0 +0.00% BenchmarkMetricVecWithLabelValues10Keys1000ValueCardinality-4 0 0 +0.00% ``` Signed-off-by: Stephen J Day <stephen.day@docker.com>
2016-08-12 03:06:03 +03:00
}
func inlineLabelValues(lvs []string, curry []curriedLabelValue) []string {
labelValues := make([]string, len(lvs)+len(curry))
var iCurry, iLVs int
for i := range labelValues {
if iCurry < len(curry) && curry[iCurry].index == i {
labelValues[i] = curry[iCurry].value
iCurry++
continue
}
labelValues[i] = lvs[iLVs]
iLVs++
}
return labelValues
}