Add support for smallint to PrefixSort

[kevincmchen]

According to the benchmark, for data sets larger than 0.5k, PrefixSort outperforms std::sort with performance improvements ranging from approximately 250% to over 500%. Here's a summary of the benchmark results:

Dataset Size	PrefixSort Improvement (No Payload)	PrefixSort Improvement(With Payload)
0.5k	248.97% - 287.43%	249.71% - 289.74%
1k	214.44% - 310.92%	215.03% - 315.14%
10k	216.21% - 255.38%	217.88% - 256.88%
100k	279.81% - 318.26%	284.89% - 295.21%
1000k	304.36% - 351.31%	454.04% - 514.28%

follow-up #8350
Part of #6766

[netlify]

✅ Deploy Preview for meta-velox canceled.

Name	Link
🔨 Latest commit	6e422a9
🔍 Latest deploy log	https://app.netlify.com/sites/meta-velox/deploys/66e23567296cb10008660ac7

[kevincmchen]

@mbasmanova would you please take a look ?

[mbasmanova]

@kevincmchen Thank you for extending prefix-sort. I noticed you added a benchmark. What did it show?

[mbasmanova]

Why do we need support for uint16_t?

[kevincmchen]

I just followed the logic of int32/uint32 to support int16/uint16. Its only a specific explanation focused on the function itself.

[mbasmanova]

@skadilover Do we need support for unsigned integer types? Is there any use case for these?

[skadilover]

They are used internal, singed ints are always converted to usigned ints

template <>
FOLLY_ALWAYS_INLINE void PrefixSortEncoder::encodeNoNulls(
    int64_t value,
    char* dest) const {
  encodeNoNulls((uint64_t)(value ^ (1ull << 63)), dest);
}

[mbasmanova]

@kevincmchen CI is red. Would you take a look?

[kevincmchen]

@kevincmchen Thank you for extending prefix-sort. I noticed you added a benchmark. What did it show?

here is small int benchmark: prefixSort vs std:sort

[mbasmanova]

here is small int benchmark: prefixSort vs std:sort

This is a lot of numbers. Would you summary the results and add the summary to the PR description?

[kevincmchen]

here is small int benchmark: prefixSort vs std:sort

This is a lot of numbers. Would you summary the results and add the summary to the PR description?

here is small int benchmark: prefixSort vs std:sort

This is a lot of numbers. Would you summary the results and add the summary to the PR description?

Here's a table summarizing the benchmark results:

Dataset Size	PrefixSort Improvement (No Payload)	PrefixSort Improvement(With Payload)
0.5k	248.97% - 287.43%	249.71% - 289.74%
1k	214.44% - 310.92%	215.03% - 315.14%
10k	216.21% - 255.38%	217.88% - 256.88%
100k	279.81% - 318.26%	284.89% - 295.21%
1000k	304.36% - 351.31%	454.04% - 514.28%

@mbasmanova

[mbasmanova]

Thanks.

[facebook-github-bot]

@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[kevincmchen]

@kevincmchen CI is red. Would you take a look?

this is due to hadoop upgrade and #10947 is working on this issue.

[kevincmchen]

Thanks.

Thanks @mbasmanova

[mbasmanova]

@kevincmchen Would you rebase to resolve CI failures?

[kevincmchen]

@kevincmchen Would you rebase to resolve CI failures?

Thanks for your reminder. i will rebase and solve the CI failures.

[facebook-github-bot]

@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[kevincmchen]

@mbasmanova there are some facebook internal CI failures , could you please help me check it ?

[mbasmanova]

@kevincmchen Seeing merge conflicts. Would you rebase once again?

[facebook-github-bot]

@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[kevincmchen]

@mbasmanova this pr #10965 should be merged first, or it will lead failure when building PyVelox.

[kevincmchen]

@mbasmanova CI is all passed. would you please help me merge this pr?

[facebook-github-bot]

@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[facebook-github-bot]

@mbasmanova merged this pull request in 98bbb73.

[conbench-facebook]

Conbench analyzed the 1 benchmark run on commit 98bbb73e.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details.