VictoriaMetrics/lib/streamaggr/dedup_test.go
Aliaksandr Valialkin 28a9e92b5e
lib/streamaggr: huge pile of changes
- Reduce memory usage by up to 5x when de-duplicating samples across big number of time series.
- Reduce memory usage by up to 5x when aggregating across big number of output time series.
- Add lib/promutils.LabelsCompressor, which is going to be used by other VictoriaMetrics components
  for reducing memory usage for marshaled []prompbmarshal.Label.
- Add `dedup_interval` option at aggregation config, which allows setting individual
  deduplication intervals per each aggregation.
- Add `keep_metric_names` option at aggregation config, which allows keeping the original
  metric names in the output samples.
- Add `unique_samples` output, which counts the number of unique sample values.
- Add `increase_prometheus` and `total_prometheus` outputs, which ignore the first sample
  per each newly encountered time series.
- Use 64-bit hashes instead of marshaled labels as map keys when calculating `count_series` output.
  This makes obsolete https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5579
- Expose various metrics, which may help debugging stream aggregation:
  - vm_streamaggr_dedup_state_size_bytes - the size of data structures responsible for deduplication
  - vm_streamaggr_dedup_state_items_count - the number of items in the deduplication data structures
  - vm_streamaggr_labels_compressor_size_bytes - the size of labels compressor data structures
  - vm_streamaggr_labels_compressor_items_count - the number of entries in the labels compressor
  - vm_streamaggr_flush_duration_seconds - a histogram, which shows the duration of stream aggregation flushes
  - vm_streamaggr_dedup_flush_duration_seconds - a histogram, which shows the duration of deduplication flushes
  - vm_streamaggr_flush_timeouts_total - counter for timed out stream aggregation flushes,
    which took longer than the configured interval
  - vm_streamaggr_dedup_flush_timeouts_total - counter for timed out deduplication flushes,
    which took longer than the configured dedup_interval
- Actualize docs/stream-aggregation.md

The memory usage reduction increases CPU usage during stream aggregation by up to 30%.

This commit is based on https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5850
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5898
2024-03-02 02:42:50 +02:00

88 lines
2.2 KiB
Go

package streamaggr
import (
"fmt"
"reflect"
"strings"
"sync"
"sync/atomic"
"testing"
)
func TestDedupAggrSerial(t *testing.T) {
da := newDedupAggr()
const seriesCount = 100_000
expectedSamplesMap := make(map[string]pushSample)
for i := 0; i < 2; i++ {
samples := make([]pushSample, seriesCount)
for j := range samples {
sample := &samples[j]
sample.key = fmt.Sprintf("key_%d", j)
sample.value = float64(i + j)
expectedSamplesMap[sample.key] = *sample
}
da.pushSamples(samples)
}
if n := da.sizeBytes(); n > 4_200_000 {
t.Fatalf("too big dedupAggr state before flush: %d bytes; it shouldn't exceed 4_200_000 bytes", n)
}
if n := da.itemsCount(); n != seriesCount {
t.Fatalf("unexpected itemsCount; got %d; want %d", n, seriesCount)
}
flushedSamplesMap := make(map[string]pushSample)
flushSamples := func(samples []pushSample) {
for _, sample := range samples {
sample.key = strings.Clone(sample.key)
flushedSamplesMap[sample.key] = sample
}
}
da.flush(flushSamples)
if !reflect.DeepEqual(expectedSamplesMap, flushedSamplesMap) {
t.Fatalf("unexpected samples;\ngot\n%v\nwant\n%v", flushedSamplesMap, expectedSamplesMap)
}
if n := da.sizeBytes(); n > 17_000 {
t.Fatalf("too big dedupAggr state after flush; %d bytes; it shouldn't exceed 17_000 bytes", n)
}
if n := da.itemsCount(); n != 0 {
t.Fatalf("unexpected non-zero itemsCount after flush; got %d", n)
}
}
func TestDedupAggrConcurrent(t *testing.T) {
const concurrency = 5
const seriesCount = 10_000
da := newDedupAggr()
var samplesFlushed atomic.Int64
flushSamples := func(samples []pushSample) {
samplesFlushed.Add(int64(len(samples)))
}
var wg sync.WaitGroup
for i := 0; i < concurrency; i++ {
wg.Add(1)
go func() {
defer wg.Done()
for i := 0; i < 10; i++ {
samples := make([]pushSample, seriesCount)
for j := range samples {
sample := &samples[j]
sample.key = fmt.Sprintf("key_%d", j)
sample.value = float64(i + j)
}
da.pushSamples(samples)
}
da.flush(flushSamples)
}()
}
wg.Wait()
if n := samplesFlushed.Load(); n < seriesCount {
t.Fatalf("too small number of series flushed; got %d; want at least %d", n, seriesCount)
}
}