Add native histograms

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Apply suggestions from code review

Co-authored-by: Arthur Silva Sens <arthursens2005@gmail.com>
Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Fix references and existing tests

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Review comments

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

fmt

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

validation

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Count validation

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Validation

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Review comments

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Space formatting

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Review comments

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Rename SyncMaptoMap -> SyncMapToMap

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Remove exemplars from parameters for constNativeHistogram function

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Update prometheus/histogram.go

Co-authored-by: George Krajcsovits <krajorama@users.noreply.github.com>
Signed-off-by: Shivanth MP <shivanth.metchem@deliveryhero.com>

Update prometheus/histogram.go

Co-authored-by: George Krajcsovits <krajorama@users.noreply.github.com>
Signed-off-by: Shivanth MP <shivanth.metchem@deliveryhero.com>

Update prometheus/histogram.go

Co-authored-by: George Krajcsovits <krajorama@users.noreply.github.com>
Signed-off-by: Shivanth MP <shivanth.metchem@deliveryhero.com>

Update prometheus/histogram.go

Co-authored-by: Björn Rabenstein <github@rabenste.in>
Signed-off-by: Shivanth MP <shivanth.metchem@deliveryhero.com>

Review comments

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>

Lint fix

Signed-off-by: Shivanth <shivanth.metchem@deliveryhero.com>
This commit is contained in:
Shivanth 2024-10-16 14:30:59 +02:00
parent 13851e9287
commit ae8497940d
2 changed files with 744 additions and 4 deletions

View File

@ -14,6 +14,7 @@
package prometheus package prometheus
import ( import (
"errors"
"fmt" "fmt"
"math" "math"
"runtime" "runtime"
@ -28,6 +29,11 @@ import (
"google.golang.org/protobuf/types/known/timestamppb" "google.golang.org/protobuf/types/known/timestamppb"
) )
const (
nativeHistogramSchemaMaximum = 8
nativeHistogramSchemaMinimum = -4
)
// nativeHistogramBounds for the frac of observed values. Only relevant for // nativeHistogramBounds for the frac of observed values. Only relevant for
// schema > 0. The position in the slice is the schema. (0 is never used, just // schema > 0. The position in the slice is the schema. (0 is never used, just
// here for convenience of using the schema directly as the index.) // here for convenience of using the schema directly as the index.)
@ -1460,9 +1466,9 @@ func pickSchema(bucketFactor float64) int32 {
floor := math.Floor(math.Log2(math.Log2(bucketFactor))) floor := math.Floor(math.Log2(math.Log2(bucketFactor)))
switch { switch {
case floor <= -8: case floor <= -8:
return 8 return nativeHistogramSchemaMaximum
case floor >= 4: case floor >= 4:
return -4 return nativeHistogramSchemaMinimum
default: default:
return -int32(floor) return -int32(floor)
} }
@ -1851,3 +1857,196 @@ func (n *nativeExemplars) addExemplar(e *dto.Exemplar) {
n.exemplars = append(n.exemplars[:nIdx], append([]*dto.Exemplar{e}, append(n.exemplars[nIdx:rIdx], n.exemplars[rIdx+1:]...)...)...) n.exemplars = append(n.exemplars[:nIdx], append([]*dto.Exemplar{e}, append(n.exemplars[nIdx:rIdx], n.exemplars[rIdx+1:]...)...)...)
} }
} }
type constNativeHistogram struct {
desc *Desc
dto.Histogram
labelPairs []*dto.LabelPair
}
func validateCount(sum float64, count uint64, negativeBuckets, positiveBuckets map[int]int64, zeroBucket uint64) error {
var bucketPopulationSum int64
for _, v := range positiveBuckets {
bucketPopulationSum += v
}
for _, v := range negativeBuckets {
bucketPopulationSum += v
}
bucketPopulationSum += int64(zeroBucket)
// If the sum of observations is NaN, the number of observations must be greater or equal to the sum of all bucket counts.
// Otherwise, the number of observations must be equal to the sum of all bucket counts .
if math.IsNaN(sum) && bucketPopulationSum > int64(count) ||
!math.IsNaN(sum) && bucketPopulationSum != int64(count) {
return errors.New("the sum of all bucket populations exceeds the count of observations")
}
return nil
}
// NewConstNativeHistogram returns a metric representing a Prometheus native histogram with
// fixed values for the count, sum, and positive/negative/zero bucket counts. As those parameters
// cannot be changed, the returned value does not implement the Histogram
// interface (but only the Metric interface). Users of this package will not
// have much use for it in regular operations. However, when implementing custom
// OpenTelemetry Collectors, it is useful as a throw-away metric that is generated on the fly
// to send it to Prometheus in the Collect method.
//
// zeroBucket counts all (positive and negative)
// observations in the zero bucket (with an absolute value less or equal
// the current threshold).
// positiveBuckets and negativeBuckets are separate maps for negative and positive
// observations. The map's value is an int64, counting observations in
// that bucket. The map's key is the
// index of the bucket according to the used
// Schema. Index 0 is for an upper bound of 1 in positive buckets and for a lower bound of -1 in negative buckets.
// NewConstNativeHistogram returns an error if
// - the length of labelValues is not consistent with the variable labels in Desc or if Desc is invalid.
// - the schema passed is not between 8 and -4
// - the sum of counts in all buckets including the zero bucket does not equal the count if sum is not NaN (or exceeds the count if sum is NaN)
//
// See https://opentelemetry.io/docs/specs/otel/compatibility/prometheus_and_openmetrics/#exponential-histograms for more details about the conversion from OTel to Prometheus.
func NewConstNativeHistogram(
desc *Desc,
count uint64,
sum float64,
positiveBuckets, negativeBuckets map[int]int64,
zeroBucket uint64,
schema int32,
zeroThreshold float64,
createdTimestamp time.Time,
labelValues ...string,
) (Metric, error) {
if desc.err != nil {
return nil, desc.err
}
if err := validateLabelValues(labelValues, len(desc.variableLabels.names)); err != nil {
return nil, err
}
if schema > nativeHistogramSchemaMaximum || schema < nativeHistogramSchemaMinimum {
return nil, errors.New("invalid native histogram schema")
}
if err := validateCount(sum, count, negativeBuckets, positiveBuckets, zeroBucket); err != nil {
return nil, err
}
NegativeSpan, NegativeDelta := makeBucketsFromMap(negativeBuckets)
PositiveSpan, PositiveDelta := makeBucketsFromMap(positiveBuckets)
ret := &constNativeHistogram{
desc: desc,
Histogram: dto.Histogram{
CreatedTimestamp: timestamppb.New(createdTimestamp),
Schema: &schema,
ZeroThreshold: &zeroThreshold,
SampleCount: &count,
SampleSum: &sum,
NegativeSpan: NegativeSpan,
NegativeDelta: NegativeDelta,
PositiveSpan: PositiveSpan,
PositiveDelta: PositiveDelta,
ZeroCount: proto.Uint64(zeroBucket),
},
labelPairs: MakeLabelPairs(desc, labelValues),
}
if *ret.ZeroThreshold == 0 && *ret.ZeroCount == 0 && len(ret.PositiveSpan) == 0 && len(ret.NegativeSpan) == 0 {
ret.PositiveSpan = []*dto.BucketSpan{{
Offset: proto.Int32(0),
Length: proto.Uint32(0),
}}
}
return ret, nil
}
// MustNewConstNativeHistogram is a version of NewConstNativeHistogram that panics where
// NewConstNativeHistogram would have returned an error.
func MustNewConstNativeHistogram(
desc *Desc,
count uint64,
sum float64,
positiveBuckets, negativeBuckets map[int]int64,
zeroBucket uint64,
nativeHistogramSchema int32,
nativeHistogramZeroThreshold float64,
createdTimestamp time.Time,
labelValues ...string,
) Metric {
nativehistogram, err := NewConstNativeHistogram(desc,
count,
sum,
positiveBuckets,
negativeBuckets,
zeroBucket,
nativeHistogramSchema,
nativeHistogramZeroThreshold,
createdTimestamp,
labelValues...)
if err != nil {
panic(err)
}
return nativehistogram
}
func (h *constNativeHistogram) Desc() *Desc {
return h.desc
}
func (h *constNativeHistogram) Write(out *dto.Metric) error {
out.Histogram = &h.Histogram
out.Label = h.labelPairs
return nil
}
func makeBucketsFromMap(buckets map[int]int64) ([]*dto.BucketSpan, []int64) {
if len(buckets) == 0 {
return nil, nil
}
var ii []int
for k := range buckets {
ii = append(ii, k)
}
sort.Ints(ii)
var (
spans []*dto.BucketSpan
deltas []int64
prevCount int64
nextI int
)
appendDelta := func(count int64) {
*spans[len(spans)-1].Length++
deltas = append(deltas, count-prevCount)
prevCount = count
}
for n, i := range ii {
count := buckets[i]
// Multiple spans with only small gaps in between are probably
// encoded more efficiently as one larger span with a few empty
// buckets. Needs some research to find the sweet spot. For now,
// we assume that gaps of one or two buckets should not create
// a new span.
iDelta := int32(i - nextI)
if n == 0 || iDelta > 2 {
// We have to create a new span, either because we are
// at the very beginning, or because we have found a gap
// of more than two buckets.
spans = append(spans, &dto.BucketSpan{
Offset: proto.Int32(iDelta),
Length: proto.Uint32(0),
})
} else {
// We have found a small gap (or no gap at all).
// Insert empty buckets as needed.
for j := int32(0); j < iDelta; j++ {
appendDelta(0)
}
}
appendDelta(count)
nextI = i + 1
}
return spans, deltas
}

View File

@ -25,11 +25,11 @@ import (
"testing/quick" "testing/quick"
"time" "time"
"github.com/prometheus/client_golang/prometheus/internal"
dto "github.com/prometheus/client_model/go" dto "github.com/prometheus/client_model/go"
"google.golang.org/protobuf/proto" "google.golang.org/protobuf/proto"
"google.golang.org/protobuf/types/known/timestamppb" "google.golang.org/protobuf/types/known/timestamppb"
"github.com/prometheus/client_golang/prometheus/internal"
) )
func benchmarkHistogramObserve(w int, b *testing.B) { func benchmarkHistogramObserve(w int, b *testing.B) {
@ -1543,3 +1543,544 @@ func TestFindBucket(t *testing.T) {
} }
} }
} }
func syncMapToMap(syncmap *sync.Map) (m map[int]int64) {
m = map[int]int64{}
syncmap.Range(func(key, value any) bool {
m[key.(int)] = *value.(*int64)
return true
})
return m
}
func TestConstNativeHistogram(t *testing.T) {
now := time.Now()
scenarios := []struct {
name string
observations []float64 // With simulated interval of 1m.
factor float64
zeroThreshold float64
maxBuckets uint32
minResetDuration time.Duration
maxZeroThreshold float64
want *dto.Histogram
}{
{
name: "no observations",
factor: 1.1,
want: &dto.Histogram{
SampleCount: proto.Uint64(0),
SampleSum: proto.Float64(0),
Schema: proto.Int32(3),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "no observations and zero threshold of zero resulting in no-op span",
factor: 1.1,
zeroThreshold: NativeHistogramZeroThresholdZero,
want: &dto.Histogram{
SampleCount: proto.Uint64(0),
SampleSum: proto.Float64(0),
Schema: proto.Int32(3),
ZeroThreshold: proto.Float64(0),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(0)},
},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "factor 1.1 results in schema 3",
observations: []float64{0, 1, 2, 3},
factor: 1.1,
want: &dto.Histogram{
SampleCount: proto.Uint64(4),
SampleSum: proto.Float64(6),
Schema: proto.Int32(3),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(1)},
{Offset: proto.Int32(7), Length: proto.Uint32(1)},
{Offset: proto.Int32(4), Length: proto.Uint32(1)},
},
PositiveDelta: []int64{1, 0, 0},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "factor 1.2 results in schema 2",
observations: []float64{0, 1, 1.2, 1.4, 1.8, 2},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(6),
SampleSum: proto.Float64(7.4),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "factor 4 results in schema -1",
observations: []float64{
0.0156251, 0.0625, // Bucket -2: (0.015625, 0.0625)
0.1, 0.25, // Bucket -1: (0.0625, 0.25]
0.5, 1, // Bucket 0: (0.25, 1]
1.5, 2, 3, 3.5, // Bucket 1: (1, 4]
5, 6, 7, // Bucket 2: (4, 16]
33.33, // Bucket 3: (16, 64]
},
factor: 4,
want: &dto.Histogram{
SampleCount: proto.Uint64(14),
SampleSum: proto.Float64(63.2581251),
Schema: proto.Int32(-1),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(-2), Length: proto.Uint32(6)},
},
PositiveDelta: []int64{2, 0, 0, 2, -1, -2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "factor 17 results in schema -2",
observations: []float64{
0.0156251, 0.0625, // Bucket -1: (0.015625, 0.0625]
0.1, 0.25, 0.5, 1, // Bucket 0: (0.0625, 1]
1.5, 2, 3, 3.5, 5, 6, 7, // Bucket 1: (1, 16]
33.33, // Bucket 2: (16, 256]
},
factor: 17,
want: &dto.Histogram{
SampleCount: proto.Uint64(14),
SampleSum: proto.Float64(63.2581251),
Schema: proto.Int32(-2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(-1), Length: proto.Uint32(4)},
},
PositiveDelta: []int64{2, 2, 3, -6},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "negative buckets",
observations: []float64{0, -1, -1.2, -1.4, -1.8, -2},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(6),
SampleSum: proto.Float64(-7.4),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
NegativeDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "negative and positive buckets",
observations: []float64{0, -1, -1.2, -1.4, -1.8, -2, 1, 1.2, 1.4, 1.8, 2},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(11),
SampleSum: proto.Float64(0),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
NegativeDelta: []int64{1, -1, 2, -2, 2},
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "wide zero bucket",
observations: []float64{0, -1, -1.2, -1.4, -1.8, -2, 1, 1.2, 1.4, 1.8, 2},
factor: 1.2,
zeroThreshold: 1.4,
want: &dto.Histogram{
SampleCount: proto.Uint64(11),
SampleSum: proto.Float64(0),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(1.4),
ZeroCount: proto.Uint64(7),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(4), Length: proto.Uint32(1)},
},
NegativeDelta: []int64{2},
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(4), Length: proto.Uint32(1)},
},
PositiveDelta: []int64{2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "NaN observation",
observations: []float64{0, 1, 1.2, 1.4, 1.8, 2, math.NaN()},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(7),
SampleSum: proto.Float64(math.NaN()),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "+Inf observation",
observations: []float64{0, 1, 1.2, 1.4, 1.8, 2, math.Inf(+1)},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(7),
SampleSum: proto.Float64(math.Inf(+1)),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
{Offset: proto.Int32(4092), Length: proto.Uint32(1)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2, -1},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "-Inf observation",
observations: []float64{0, 1, 1.2, 1.4, 1.8, 2, math.Inf(-1)},
factor: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(7),
SampleSum: proto.Float64(math.Inf(-1)),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(4097), Length: proto.Uint32(1)},
},
NegativeDelta: []int64{1},
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "limited buckets but nothing triggered",
observations: []float64{0, 1, 1.2, 1.4, 1.8, 2},
factor: 1.2,
maxBuckets: 4,
want: &dto.Histogram{
SampleCount: proto.Uint64(6),
SampleSum: proto.Float64(7.4),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by halving resolution",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 3},
factor: 1.2,
maxBuckets: 4,
want: &dto.Histogram{
SampleCount: proto.Uint64(8),
SampleSum: proto.Float64(11.5),
Schema: proto.Int32(1),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
PositiveDelta: []int64{1, 2, -1, -2, 1},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by widening the zero bucket",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 3},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(8),
SampleSum: proto.Float64(11.5),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(1),
ZeroCount: proto.Uint64(2),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(1), Length: proto.Uint32(7)},
},
PositiveDelta: []int64{1, 1, -2, 2, -2, 0, 1},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by widening the zero bucket twice",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 3, 4},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(9),
SampleSum: proto.Float64(15.5),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(1.189207115002721),
ZeroCount: proto.Uint64(3),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(2), Length: proto.Uint32(7)},
},
PositiveDelta: []int64{2, -2, 2, -2, 0, 1, 0},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by reset",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 3, 4},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
minResetDuration: 5 * time.Minute,
want: &dto.Histogram{
SampleCount: proto.Uint64(2),
SampleSum: proto.Float64(7),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(7), Length: proto.Uint32(2)},
},
PositiveDelta: []int64{1, 0},
CreatedTimestamp: timestamppb.New(now.Add(8 * time.Minute)), // We expect reset to happen after 8 observations.
},
},
{
name: "limited buckets but nothing triggered, negative observations",
observations: []float64{0, -1, -1.2, -1.4, -1.8, -2},
factor: 1.2,
maxBuckets: 4,
want: &dto.Histogram{
SampleCount: proto.Uint64(6),
SampleSum: proto.Float64(-7.4),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
NegativeDelta: []int64{1, -1, 2, -2, 2},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by halving resolution, negative observations",
observations: []float64{0, -1, -1.1, -1.2, -1.4, -1.8, -2, -3},
factor: 1.2,
maxBuckets: 4,
want: &dto.Histogram{
SampleCount: proto.Uint64(8),
SampleSum: proto.Float64(-11.5),
Schema: proto.Int32(1),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(1),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(0), Length: proto.Uint32(5)},
},
NegativeDelta: []int64{1, 2, -1, -2, 1},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by widening the zero bucket, negative observations",
observations: []float64{0, -1, -1.1, -1.2, -1.4, -1.8, -2, -3},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(8),
SampleSum: proto.Float64(-11.5),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(1),
ZeroCount: proto.Uint64(2),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(1), Length: proto.Uint32(7)},
},
NegativeDelta: []int64{1, 1, -2, 2, -2, 0, 1},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by widening the zero bucket twice, negative observations",
observations: []float64{0, -1, -1.1, -1.2, -1.4, -1.8, -2, -3, -4},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
want: &dto.Histogram{
SampleCount: proto.Uint64(9),
SampleSum: proto.Float64(-15.5),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(1.189207115002721),
ZeroCount: proto.Uint64(3),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(2), Length: proto.Uint32(7)},
},
NegativeDelta: []int64{2, -2, 2, -2, 0, 1, 0},
CreatedTimestamp: timestamppb.New(now),
},
},
{
name: "buckets limited by reset, negative observations",
observations: []float64{0, -1, -1.1, -1.2, -1.4, -1.8, -2, -3, -4},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
minResetDuration: 5 * time.Minute,
want: &dto.Histogram{
SampleCount: proto.Uint64(2),
SampleSum: proto.Float64(-7),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
NegativeSpan: []*dto.BucketSpan{
{Offset: proto.Int32(7), Length: proto.Uint32(2)},
},
NegativeDelta: []int64{1, 0},
CreatedTimestamp: timestamppb.New(now.Add(8 * time.Minute)), // We expect reset to happen after 8 observations.
},
},
{
name: "buckets limited by halving resolution, then reset",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 5, 5.1, 3, 4},
factor: 1.2,
maxBuckets: 4,
minResetDuration: 9 * time.Minute,
want: &dto.Histogram{
SampleCount: proto.Uint64(3),
SampleSum: proto.Float64(12.1),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(7), Length: proto.Uint32(4)},
},
PositiveDelta: []int64{1, 0, -1, 1},
CreatedTimestamp: timestamppb.New(now.Add(9 * time.Minute)), // We expect reset to happen after 8 minutes.
},
},
{
name: "buckets limited by widening the zero bucket, then reset",
observations: []float64{0, 1, 1.1, 1.2, 1.4, 1.8, 2, 5, 5.1, 3, 4},
factor: 1.2,
maxBuckets: 4,
maxZeroThreshold: 1.2,
minResetDuration: 9 * time.Minute,
want: &dto.Histogram{
SampleCount: proto.Uint64(3),
SampleSum: proto.Float64(12.1),
Schema: proto.Int32(2),
ZeroThreshold: proto.Float64(2.938735877055719e-39),
ZeroCount: proto.Uint64(0),
PositiveSpan: []*dto.BucketSpan{
{Offset: proto.Int32(7), Length: proto.Uint32(4)},
},
PositiveDelta: []int64{1, 0, -1, 1},
CreatedTimestamp: timestamppb.New(now.Add(9 * time.Minute)), // We expect reset to happen after 8 minutes.
},
},
}
for _, s := range scenarios {
t.Run(s.name, func(t *testing.T) {
var (
ts = now
funcToCall func()
whenToCall time.Duration
)
his := NewHistogram(HistogramOpts{
Name: "name",
Help: "help",
NativeHistogramBucketFactor: s.factor,
NativeHistogramZeroThreshold: s.zeroThreshold,
NativeHistogramMaxBucketNumber: s.maxBuckets,
NativeHistogramMinResetDuration: s.minResetDuration,
NativeHistogramMaxZeroThreshold: s.maxZeroThreshold,
now: func() time.Time { return ts },
afterFunc: func(d time.Duration, f func()) *time.Timer {
funcToCall = f
whenToCall = d
return nil
},
})
ts = ts.Add(time.Minute)
for _, o := range s.observations {
his.Observe(o)
ts = ts.Add(time.Minute)
whenToCall -= time.Minute
if funcToCall != nil && whenToCall <= 0 {
funcToCall()
funcToCall = nil
}
}
_his := his.(*histogram)
n := atomic.LoadUint64(&_his.countAndHotIdx)
hotIdx := n >> 63
cold := _his.counts[hotIdx]
consthist, err := NewConstNativeHistogram(_his.Desc(),
cold.count,
math.Float64frombits(cold.sumBits),
syncMapToMap(&cold.nativeHistogramBucketsPositive),
syncMapToMap(&cold.nativeHistogramBucketsNegative),
cold.nativeHistogramZeroBucket,
cold.nativeHistogramSchema,
math.Float64frombits(cold.nativeHistogramZeroThresholdBits),
_his.lastResetTime,
)
if err != nil {
t.Fatal("unexpected error writing metric", err)
}
m2 := &dto.Metric{}
if err := consthist.Write(m2); err != nil {
t.Fatal("unexpected error writing metric", err)
}
got := m2.Histogram
if !proto.Equal(s.want, got) {
t.Errorf("want histogram %q, got %q", s.want, got)
}
})
}
}