You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "fallintoplace (via GitHub)" <gi...@apache.org> on 2026/08/27 10:29:21 UTC
[PR] perf(parquet): reuse boolean spaced scratch [arrow-go]
fallintoplace opened a new pull request, #1231:
URL: https://github.com/apache/arrow-go/pull/1231
## What
- Reuse a `[]bool` scratch buffer in `PutSpaced` for Plain and RLE boolean encoders.
- Reuse the same scratch for RLE `PutSpacedBitmap` writes.
- Keep the scratch capacity across repeated page writes.
## Benchmark
Apple M1 Pro, Go 1.26.3. The same encoder is reused, and scratch growth happens before the timed loop.
Command: `go test -vet=off ./parquet/internal/encoding -run "^$" -bench "^Benchmark(Plain|Rle)BooleanEncoderPutSpaced$" -benchmem -benchtime=300ms -count=3 -cpu=1`
All-valid input, before vs after:
| Encoder | Values | Before | After |
| --- | ---: | ---: | ---: |
| Plain | 1,024 | 1,024 B/op, 1 alloc | 0 B/op, 0 allocs |
| Plain | 65,536 | 65,539 B/op, 1 alloc | 0 B/op, 0 allocs |
| Plain | 1,048,576 | 1,048,620 B/op, 1 alloc | 0 B/op, 0 allocs |
| RLE | 1,024 | 1,232 B/op, 4 allocs | 208 B/op, 3 allocs |
| RLE | 65,536 | 65,747 B/op, 4 allocs | 208 B/op, 3 allocs |
## Tests
- `go test ./parquet/internal/encoding -count=1`
- `PARQUET_TEST_DATA=parquet-testing/data go test ./... -count=1`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] perf(parquet): reuse boolean spaced scratch [arrow-go]
Posted by "zeroshade (via GitHub)" <gi...@apache.org>.
zeroshade merged PR #1231:
URL: https://github.com/apache/arrow-go/pull/1231
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org