perf: optimize memory usage for client.List func #369
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
If there is a predicate condition, we should not alloc a very large
slice
```
goos: linux
goarch: amd64
pkg: github.com/ovn-org/libovsdb/client
cpu: Intel(R) Core(TM) i7-1065G7 CPU @ 1.30GHz
│ bench.out.main │ bench.out.new │
│ sec/op │ sec/op vs base │
APIList/predicate_returns_none-4 3.419m ± 41% 3.238m ± 1% -5.29% (p=0.001 n=10)
APIList/predicate_returns_all-4 67.92m ± 3% 68.95m ± 1% ~ (p=0.123 n=10)
APIList/predicate_on_an_arbitrary_condition-4 11.88m ± 2% 12.10m ± 11% ~ (p=0.075 n=10)
APIList/predicate_matches_name-4 4.073m ± 4% 5.006m ± 17% +22.91% (p=0.000 n=10)
APIList/by_index,_no_predicate-4 32.644µ ± 4% 7.957µ ± 3% -75.62% (p=0.000 n=10)
APIListMultiple/multiple_results_one_at_a_time_with_Get-4 6.073m ± 4% 6.021m ± 11% ~ (p=0.853 n=10)
APIListMultiple/multiple_results_in_a_batch_with_WhereAny-4 3.014m ± 2% 3.085m ± 19% +2.36% (p=0.002 n=10)
geomean 3.521m 2.962m -15.89%
│ bench.out.main │ bench.out.new │
│ B/op │ B/op vs base │
APIList/predicate_returns_none-4 908.6Ki ± 0% 828.6Ki ± 0% -8.80% (p=0.000 n=10)
APIList/predicate_returns_all-4 14.88Mi ± 0% 14.88Mi ± 0% ~ (p=0.315 n=10)
APIList/predicate_on_an_arbitrary_condition-4 2.468Mi ± 0% 2.399Mi ± 0% -2.80% (p=0.000 n=10)
APIList/predicate_matches_name-4 910.3Ki ± 0% 830.3Ki ± 0% -8.79% (p=0.000 n=10)
APIList/by_index,_no_predicate-4 81.997Ki ± 0% 2.000Ki ± 0% -97.56% (p=0.000 n=10)
APIListMultiple/multiple_results_one_at_a_time_with_Get-4 1.439Mi ± 0% 1.439Mi ± 0% ~ (p=0.197 n=10)
APIListMultiple/multiple_results_in_a_batch_with_WhereAny-4 712.7Ki ± 0% 712.7Ki ± 0% ~ (p=0.280 n=10)
geomean 1.128Mi 659.2Ki -42.93%
│ bench.out.main │ bench.out.new │
│ allocs/op │ allocs/op vs base │
APIList/predicate_returns_none-4 20.01k ± 0% 20.01k ± 0% -0.00% (p=0.000 n=10)
APIList/predicate_returns_all-4 230.1k ± 0% 230.1k ± 0% +0.00% (p=0.048 n=10)
APIList/predicate_on_an_arbitrary_condition-4 43.36k ± 0% 43.36k ± 0% ~ (p=1.000 n=10)
APIList/predicate_matches_name-4 20.03k ± 0% 20.03k ± 0% ~ (p=1.000 n=10) ¹
APIList/by_index,_no_predicate-4 32.00 ± 0% 32.00 ± 0% ~ (p=1.000 n=10) ¹
APIListMultiple/multiple_results_one_at_a_time_with_Get-4 22.02k ± 0% 22.01k ± 0% ~ (p=0.170 n=10)
APIListMultiple/multiple_results_in_a_batch_with_WhereAny-4 11.51k ± 0% 11.51k ± 0% ~ (p=1.000 n=10) ¹
geomean 11.83k 11.83k -0.00%
¹ all samples are equal
```
Signed-off-by: Yan Zhu <[email protected]>
halfcrazy
force-pushed
the
perf/list-mem
branch
from
bbf4840
to
219c10b
Compare
|
ping |
|
ping @jcaamano |
jcaamano
reviewed
Jan 2, 2024
client/api.go
Outdated
Comment on lines
156
to
173
| if resultVal.IsNil() || resultVal.Cap() == 0 { | ||
| resultVal.Set(reflect.MakeSlice(resultVal.Type(), 0, len(rows))) | ||
| } | ||
| } else { | ||
| rows = tableCache.Rows() | ||
| // If given a null slice, fill it in the cache table completely, if not, just up to | ||
| // its capability. | ||
| if resultVal.IsNil() || resultVal.Cap() == 0 { | ||
| resultVal.Set(reflect.MakeSlice(resultVal.Type(), 0, tableCache.Len())) | ||
| } | ||
| } | ||
| i := resultVal.Len() | ||
| maxCap := resultVal.Cap() | ||
|
|
||
| for _, row := range rows { | ||
| if i >= resultVal.Cap() { | ||
| if i >= maxCap { | ||
| break | ||
| } |
There was a problem hiding this comment.
@halfcrazy sorry this took so long. Would this work:
var rows map[string]model.Model
if a.cond != nil {
rows, err = a.cond.Matches()
if err != nil {
return err
}
} else {
rows = tableCache.Rows()
}
// If given a null slice, fill it in with all the resulting rows, if not, just up to
// its capacity.
if resultVal.IsNil() || resultVal.Cap() == 0 {
resultVal.Set(reflect.MakeSlice(resultVal.Type(), 0, len(rows)))
}
i := resultVal.Len()
cap := resultVal.Cap()
for _, row := range rows {
if i >= cap {
break
}
...
Signed-off-by: Yan Zhu <[email protected]>
halfcrazy
force-pushed
the
perf/list-mem
branch
from
518c281
to
776d5b2
Compare
jcaamano
approved these changes
Jan 2, 2024
Pull Request Test Coverage Report for Build 6441536993Warning: This coverage report may be inaccurate.This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.
Details
💛 - Coveralls |
1 similar comment
Pull Request Test Coverage Report for Build 6441536993Warning: This coverage report may be inaccurate.This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.
Details
💛 - Coveralls |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix #367
If there is a predicate condition, we should not alloc a very large slice
main
perf