1
0
mirror of https://github.com/open-telemetry/opentelemetry-go.git synced 2025-08-10 22:31:50 +02:00

Add benchmark for map access using attribute Equivalent (#7123)

I am looking into I am looking into
https://promlabs.com/blog/2025/07/17/why-i-recommend-native-prometheus-instrumentation-over-opentelemetry/#comparing-counter-increment-performance,
and was trying to figure out why incrementing a counter with 10
attributes was so much slower than incrementing a counter with no
attributes, or 1 attribute:

```
$ go test -run=xxxxxMatchNothingxxxxx -cpu=1 -test.benchtime=1s -bench=BenchmarkSyncMeasure/NoView/Int64Counter/Attributes
goos: linux
goarch: amd64
pkg: go.opentelemetry.io/otel/sdk/metric
cpu: Intel(R) Xeon(R) CPU @ 2.20GHz
BenchmarkSyncMeasure/NoView/Int64Counter/Attributes/0         	 9905773	       121.3 ns/op
BenchmarkSyncMeasure/NoView/Int64Counter/Attributes/1         	 4079145	       296.5 ns/op
BenchmarkSyncMeasure/NoView/Int64Counter/Attributes/10        	  781627	      1531 ns/op
```

Looking at the profile, most of the time is spent in
"runtime.mapKeyError2" within "runtime.mapaccess2". My best guess is
that whatever we are using for Equivalent() is not very performant when
used as a map key. This seems like a good opportunity to greatly improve
the performance of our metrics (and probably other signals) API + SDK.

To start, i'm adding a simple benchmark within the attribute package to
isolate the issue. Results:

```
$ go test     -run '^$' -bench '^BenchmarkEquivalentMapAccess'     -benchtime .1s -cpu 1 -benchmem 
goos: linux
goarch: amd64
pkg: go.opentelemetry.io/otel/attribute
cpu: Intel(R) Xeon(R) CPU @ 2.20GHz
BenchmarkEquivalentMapAccess/Empty         	 2220508	        53.58 ns/op	       0 B/op	       0 allocs/op
BenchmarkEquivalentMapAccess/1_string_attribute         	  622770	       196.7 ns/op	       0 B/op	       0 allocs/op
BenchmarkEquivalentMapAccess/10_string_attributes       	   77462	      1558 ns/op	       0 B/op	       0 allocs/op
BenchmarkEquivalentMapAccess/1_int_attribute            	  602163	       197.7 ns/op	       0 B/op	       0 allocs/op
BenchmarkEquivalentMapAccess/10_int_attributes          	   76603	      1569 ns/op	       0 B/op	       0 allocs/op
``` 

This shows that it is the map lookup and storage itself that is making
the metrics API+SDK perform much worse with more attributes.

Some optimization ideas include:

* Most attribute sets are likely to be just numbers and strings. Can we
make a fast path for sets that don't include complex attributes?
* We encourage improving performance of the metrics API by re-using
attribute sets where possible. If we can lazily compute+cache a "faster"
map key, that will have a big performance improvement when attribute
sets are re-used.
* compute a uint64 hash using something like
https://github.com/gohugoio/hashstructure, or something similar to what
prometheus/client_golang does:
c79a891c6c/model/signature.go (L31)

---------

Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
Co-authored-by: Flc゛ <four_leaf_clover@foxmail.com>
This commit is contained in:
David Ashpole
2025-08-06 14:45:22 -04:00
committed by GitHub
parent 49033aa569
commit eb4f1dc4a1

View File

@@ -271,3 +271,59 @@ func BenchmarkStringSlice(b *testing.B) {
}) })
b.Run("Emit", benchmarkEmit(kv)) b.Run("Emit", benchmarkEmit(kv))
} }
// BenchmarkEquivalentMapAccess measures how expensive it is to use
// Equivalent() as a map key. This is on the hot path for making synchronous
// measurements on the metrics API/SDK. It will likely be on the hot path for
// the trace and logs API/SDK in the future.
func BenchmarkEquivalentMapAccess(b *testing.B) {
b.Run("Empty", func(b *testing.B) {
benchmarkEquivalentMapAccess(b, attribute.EmptySet())
})
b.Run("1 string attribute", func(b *testing.B) {
set := attribute.NewSet(attribute.String("string", "42"))
benchmarkEquivalentMapAccess(b, &set)
})
b.Run("10 string attributes", func(b *testing.B) {
set := attribute.NewSet(
attribute.String("a", "42"),
attribute.String("b", "42"),
attribute.String("c", "42"),
attribute.String("d", "42"),
attribute.String("e", "42"),
attribute.String("f", "42"),
attribute.String("g", "42"),
attribute.String("h", "42"),
attribute.String("i", "42"),
attribute.String("j", "42"),
)
benchmarkEquivalentMapAccess(b, &set)
})
b.Run("1 int attribute", func(b *testing.B) {
set := attribute.NewSet(attribute.Int("string", 42))
benchmarkEquivalentMapAccess(b, &set)
})
b.Run("10 int attributes", func(b *testing.B) {
set := attribute.NewSet(
attribute.Int("a", 42),
attribute.Int("b", 42),
attribute.Int("c", 42),
attribute.Int("d", 42),
attribute.Int("e", 42),
attribute.Int("f", 42),
attribute.Int("g", 42),
attribute.Int("h", 42),
attribute.Int("i", 42),
attribute.Int("j", 42),
)
benchmarkEquivalentMapAccess(b, &set)
})
}
func benchmarkEquivalentMapAccess(b *testing.B, set *attribute.Set) {
values := map[attribute.Distinct]int{}
b.ResetTimer()
for range b.N {
values[set.Equivalent()]++
}
}