GH-43627: [R] Fix summarize() performance regression (pushdown)

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.

[nealrichardson]

Rationale for this change

See #43627 (comment)

What changes are included in this PR?

An extra dplyr::select()

Are these changes tested?

Conbench should show that the performance is much better

Are there any user-facing changes?

Not slow

• GitHub Issue: [R] Possible regression in dev arrow #43627

[github-actions]

⚠️ GitHub issue #43627has been automatically assigned in GitHub to PR creator.

[jonkeane]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 4493c6c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 4 benchmarking runs that have been run so far on PR commit 4493c6c.

There were 96 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-08-13 16:34:27Z
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/0, source=cpp-micro, suite=arrow-compute-grouper-benchmark
  • GrouperWithMultiTypes (C++) with params="{utf8}"/1024/10000, source=cpp-micro, suite=arrow-compute-grouper-benchmark
• and 94 more (see the report linked below)

The full Conbench report has more details.

[nealrichardson]

Conbench confirms that this addresses the performance regression. Both of the benchmarks flagged on #41576 (comment) are back to their previous (better) results. Here's one of them: https://conbench.ursa.dev/compare/benchmark-results/066ba4064f317b748000fc9a4515588b...066bb8b26d287d6c80000b57fe949079/

[jonkeane]

Thanks for this

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ab432b1.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.