You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "llama90 (via GitHub)" <gi...@apache.org> on 2024/03/29 10:13:15 UTC
[PR] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
llama90 opened a new pull request, #40884:
URL: https://github.com/apache/arrow/pull/40884
<!--
Thanks for opening a pull request!
If this is your first pull request you can find detailed information on how
to contribute here:
* [New Contributor's Guide](https://arrow.apache.org/docs/dev/developers/guide/step_by_step/pr_lifecycle.html#reviews-and-merge-of-the-pull-request)
* [Contributing Overview](https://arrow.apache.org/docs/dev/developers/overview.html)
If this is not a [minor PR](https://github.com/apache/arrow/blob/main/CONTRIBUTING.md#Minor-Fixes). Could you open an issue for this pull request on GitHub? https://github.com/apache/arrow/issues/new/choose
Opening GitHub issues ahead of time contributes to the [Openness](http://theapacheway.com/open/#:~:text=Openness%20allows%20new%20users%20the,must%20happen%20in%20the%20open.) of the Apache Arrow project.
Then could you also rename the pull request title in the following format?
GH-${GITHUB_ISSUE_ID}: [${COMPONENT}] ${SUMMARY}
or
MINOR: [${COMPONENT}] ${SUMMARY}
In the case of PARQUET issues on JIRA the title also supports:
PARQUET-${JIRA_ISSUE_ID}: [${COMPONENT}] ${SUMMARY}
-->
### Rationale for this change
Acero benchmarks sometimes output metrics such as `items/s`, `bytes/s`, `batches/s`, and `rows/s`. However, there is inconsistency in how these metrics are presented across different benchmarks. We are undertaking an effort to standardize the output of these metrics to ensure uniformity and clarity in performance measurement across all Acero benchmarks.
<!--
Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes.
-->
### What changes are included in this PR?
`rows/s` has a similar meaning to `items/s`.
- `bytes/s` and `items/s`: aggregate
- `bytes/s` and `rows/s`: asof_join
- `batches/s` and `rows/s`: project, filter, expression
<!--
There is no need to duplicate the description in the issue here but it is sometimes worth providing a summary of the individual changes in this PR.
-->
### Are these changes tested?
Yes.
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example, are they covered by existing tests)?
-->
### Are there any user-facing changes?
No.
<!--
If there are user-facing changes then we may require documentation to be updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please uncomment the line below and explain which changes are breaking.
-->
<!-- **This PR includes breaking changes to public APIs.** -->
<!--
Please uncomment the line below (and provide explanation) if the changes fix either (a) a security vulnerability, (b) a bug that caused incorrect or invalid data to be produced, or (c) a bug that causes a crash (even when the API contract is upheld). We use this to highlight fixes to issues that may affect users without their knowledge. For this reason, fixing bugs that cause errors don't count, since those are usually obvious.
-->
<!-- **This PR contains a "Critical Fix".** -->
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "llama90 (via GitHub)" <gi...@apache.org>.
llama90 commented on code in PR #40884:
URL: https://github.com/apache/arrow/pull/40884#discussion_r1573117355
##########
cpp/src/arrow/acero/expression_benchmark.cc:
##########
@@ -66,17 +69,34 @@ static void BindAndEvaluate(benchmark::State& state, Expression expr) {
ASSIGN_OR_ABORT(auto bound, expr.Bind(*dataset_schema));
ABORT_NOT_OK(ExecuteScalarExpression(bound, input, &ctx).status());
}
+
+ state.counters["rows_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches * rows_per_batch),
+ benchmark::Counter::kIsRate);
+
+ state.counters["batches_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches), benchmark::Counter::kIsRate);
}
// A benchmark of SimplifyWithGuarantee using expressions arising from partitioning.
static void SimplifyFilterWithGuarantee(benchmark::State& state, Expression filter,
Expression guarantee) {
+ const auto rows_per_batch = static_cast<int32_t>(state.range(0));
+ const auto num_batches = 1000000 / rows_per_batch;
+
auto dataset_schema = schema({field("a", int64()), field("b", int64())});
ASSIGN_OR_ABORT(filter, filter.Bind(*dataset_schema));
for (auto _ : state) {
ABORT_NOT_OK(SimplifyWithGuarantee(filter, guarantee));
}
+
+ state.counters["rows_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches * rows_per_batch),
+ benchmark::Counter::kIsRate);
+
+ state.counters["batches_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches), benchmark::Counter::kIsRate);
Review Comment:
I rolled back all the changes related to `expressions_benchmark.cc`. As you mentioned, the metrics such as `rows_per_second` and `batches_per_second` seem unnecessary for this benchmark. While working, I was too focused on outputting these metrics.
Based on your review, it appears that those were incorrect modifications, so I have rolled back all related changes.
Thank you for the review.
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "llama90 (via GitHub)" <gi...@apache.org>.
llama90 commented on code in PR #40884:
URL: https://github.com/apache/arrow/pull/40884#discussion_r1573115746
##########
cpp/src/arrow/acero/aggregate_benchmark.cc:
##########
@@ -321,6 +321,19 @@ BENCHMARK_TEMPLATE(ReferenceSum, SumBitmapVectorizeUnroll<int64_t>)
// GroupBy
//
+int64_t CalculateBatchSize(const std::shared_ptr<RecordBatch>& batch) {
Review Comment:
I fixed it to replace CalculateBatchSize with TotalBufferSize.
Thank you for letting me know.
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #40884:
URL: https://github.com/apache/arrow/pull/40884#issuecomment-2027198761
Note that `bytes_per_second` and `items_per_second` are standard metrics with Google benchmark, so they will have to stay. For Acero, I think we should go with the convention that 1 "item" == 1 row. As for `bytes_per_second`, it's less useful than `items_per_second` IMHO, but can be kept if it already exists.
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "westonpace (via GitHub)" <gi...@apache.org>.
westonpace commented on code in PR #40884:
URL: https://github.com/apache/arrow/pull/40884#discussion_r1547750014
##########
cpp/src/arrow/acero/aggregate_benchmark.cc:
##########
@@ -321,6 +321,19 @@ BENCHMARK_TEMPLATE(ReferenceSum, SumBitmapVectorizeUnroll<int64_t>)
// GroupBy
//
+int64_t CalculateBatchSize(const std::shared_ptr<RecordBatch>& batch) {
Review Comment:
There is a method `TotalBufferSize` in `src/arrow/util/byte_size.h` that you can use.
##########
cpp/src/arrow/acero/expression_benchmark.cc:
##########
@@ -41,6 +41,9 @@ std::shared_ptr<Scalar> ninety_nine_dict =
DictionaryScalar::Make(MakeScalar(0), ArrayFromJSON(int64(), "[99]"));
static void BindAndEvaluate(benchmark::State& state, Expression expr) {
+ const auto rows_per_batch = static_cast<int32_t>(state.range(0));
+ const auto num_batches = 1000000 / rows_per_batch;
Review Comment:
Where is 1000000 coming from? It looks like each iteration is running against a batch of 4 rows.
##########
cpp/src/arrow/acero/expression_benchmark.cc:
##########
@@ -66,17 +69,34 @@ static void BindAndEvaluate(benchmark::State& state, Expression expr) {
ASSIGN_OR_ABORT(auto bound, expr.Bind(*dataset_schema));
ABORT_NOT_OK(ExecuteScalarExpression(bound, input, &ctx).status());
}
+
+ state.counters["rows_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches * rows_per_batch),
+ benchmark::Counter::kIsRate);
+
+ state.counters["batches_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches), benchmark::Counter::kIsRate);
}
// A benchmark of SimplifyWithGuarantee using expressions arising from partitioning.
static void SimplifyFilterWithGuarantee(benchmark::State& state, Expression filter,
Expression guarantee) {
+ const auto rows_per_batch = static_cast<int32_t>(state.range(0));
+ const auto num_batches = 1000000 / rows_per_batch;
+
auto dataset_schema = schema({field("a", int64()), field("b", int64())});
ASSIGN_OR_ABORT(filter, filter.Bind(*dataset_schema));
for (auto _ : state) {
ABORT_NOT_OK(SimplifyWithGuarantee(filter, guarantee));
}
+
+ state.counters["rows_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches * rows_per_batch),
+ benchmark::Counter::kIsRate);
+
+ state.counters["batches_per_second"] = benchmark::Counter(
+ static_cast<double>(state.iterations() * num_batches), benchmark::Counter::kIsRate);
Review Comment:
This benchmark doesn't have any input. It is simplifying an expression. I'm not sure there is a rows per second or batches per second that makes sense here.
##########
cpp/src/arrow/acero/expression_benchmark.cc:
##########
@@ -163,31 +183,103 @@ auto ref_only_expression = field_ref("x");
// Negative queries (partition expressions that fail the filter)
BENCHMARK_CAPTURE(SimplifyFilterWithGuarantee, negative_filter_simple_guarantee_simple,
- filter_simple_negative, guarantee);
+ filter_simple_negative, guarantee)
+ ->ArgNames({"rows_per_batch"})
+ ->RangeMultiplier(10)
+ ->Range(1000, 1000000)
Review Comment:
I'm not sure you can add `Range` / `RangeMultiplier` here since the benchmark is not looking at it then it will have no effect?
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "llama90 (via GitHub)" <gi...@apache.org>.
llama90 commented on PR #40884:
URL: https://github.com/apache/arrow/pull/40884#issuecomment-2027167389
Regarding throughput metrics in benchmarks, While the benchmarks are independent of each other, there is a mix in terms of Acero metrics. (`bytes/s`, `items/s`, `rows/s`, and `batches/s`). I am contemplating whether to unify these terms for consistency. Attached are the benchmark results for review and feedback.
I welcome opinions on this matter. cc @pitrou @kou
#### aggregate
<details><summary>result</summary>
```
Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
***WARNING*** Failed to set thread affinity. Estimated CPU frequency may be incorrect.
2024-03-29T19:05:29+09:00
Running /Users/lama/workspace/arrow-new/cpp/cmake-build-debug/debug/arrow-acero-aggregate-benchmark
Run on (10 X 24 MHz CPU s)
CPU Caches:
L1 Data 64 KiB
L1 Instruction 128 KiB
L2 Unified 4096 KiB (x10)
Load Average: 9.14, 14.98, 15.57
***WARNING*** Library was built as DEBUG. Timings may be affected.
------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations UserCounters...
------------------------------------------------------------------------------------------------------------------
SumDoublesGroupedByTinyStringSet/1048576/10000 136635 us 135963 us 5 bytes_per_second=196.772Mi/s items_per_second=7.71222M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByTinyStringSet/1048576/100 136418 us 136054 us 5 bytes_per_second=196.64Mi/s items_per_second=7.70707M/s null_percent=1 size=1.04858M
SumDoublesGroupedByTinyStringSet/1048576/10 137404 us 137105 us 5 bytes_per_second=195.132Mi/s items_per_second=7.64797M/s null_percent=10 size=1.04858M
SumDoublesGroupedByTinyStringSet/1048576/2 142042 us 138207 us 5 bytes_per_second=193.576Mi/s items_per_second=7.58699M/s null_percent=50 size=1.04858M
SumDoublesGroupedByTinyStringSet/1048576/1 133889 us 127989 us 6 bytes_per_second=209.031Mi/s items_per_second=8.1927M/s null_percent=100 size=1.04858M
SumDoublesGroupedByTinyStringSet/1048576/0 137069 us 135097 us 5 bytes_per_second=197.107Mi/s items_per_second=7.76164M/s null_percent=0 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/10000 151138 us 145978 us 5 bytes_per_second=201.813Mi/s items_per_second=7.18313M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/100 145625 us 142846 us 5 bytes_per_second=206.237Mi/s items_per_second=7.34059M/s null_percent=1 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/10 143154 us 141850 us 5 bytes_per_second=207.686Mi/s items_per_second=7.39217M/s null_percent=10 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/2 143609 us 142296 us 5 bytes_per_second=207.035Mi/s items_per_second=7.36899M/s null_percent=50 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/1 136270 us 132507 us 5 bytes_per_second=222.329Mi/s items_per_second=7.91336M/s null_percent=100 size=1.04858M
SumDoublesGroupedBySmallStringSet/1048576/0 143683 us 141840 us 5 bytes_per_second=206.818Mi/s items_per_second=7.39265M/s null_percent=0 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/10000 147634 us 145947 us 5 bytes_per_second=201.524Mi/s items_per_second=7.18465M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/100 150866 us 146933 us 5 bytes_per_second=200.172Mi/s items_per_second=7.13644M/s null_percent=1 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/10 150889 us 148290 us 5 bytes_per_second=198.339Mi/s items_per_second=7.07111M/s null_percent=10 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/2 146823 us 145840 us 5 bytes_per_second=201.671Mi/s items_per_second=7.18989M/s null_percent=50 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/1 135807 us 135116 us 5 bytes_per_second=217.677Mi/s items_per_second=7.76054M/s null_percent=100 size=1.04858M
SumDoublesGroupedByMediumStringSet/1048576/0 145426 us 144719 us 4 bytes_per_second=202.37Mi/s items_per_second=7.24559M/s null_percent=0 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/10000 50021 us 49769 us 14 bytes_per_second=323.995Mi/s items_per_second=21.0688M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/100 50427 us 50193 us 14 bytes_per_second=321.26Mi/s items_per_second=20.8909M/s null_percent=1 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/10 50205 us 49988 us 14 bytes_per_second=322.577Mi/s items_per_second=20.9766M/s null_percent=10 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/2 50506 us 50143 us 14 bytes_per_second=321.581Mi/s items_per_second=20.9117M/s null_percent=50 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/1 39904 us 39713 us 18 bytes_per_second=406.042Mi/s items_per_second=26.4041M/s null_percent=100 size=1.04858M
SumDoublesGroupedByTinyIntegerSet/1048576/0 49175 us 48990 us 14 bytes_per_second=326.598Mi/s items_per_second=21.4039M/s null_percent=0 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/10000 50194 us 49994 us 14 bytes_per_second=322.538Mi/s items_per_second=20.974M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/100 50852 us 50613 us 14 bytes_per_second=318.596Mi/s items_per_second=20.7176M/s null_percent=1 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/10 51009 us 50779 us 14 bytes_per_second=317.552Mi/s items_per_second=20.6497M/s null_percent=10 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/2 50991 us 50763 us 14 bytes_per_second=317.654Mi/s items_per_second=20.6564M/s null_percent=50 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/1 40107 us 39984 us 17 bytes_per_second=403.289Mi/s items_per_second=26.2251M/s null_percent=100 size=1.04858M
SumDoublesGroupedBySmallIntegerSet/1048576/0 49749 us 49602 us 14 bytes_per_second=322.565Mi/s items_per_second=21.1396M/s null_percent=0 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/10000 51556 us 51357 us 14 bytes_per_second=313.977Mi/s items_per_second=20.4173M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/100 51524 us 51374 us 13 bytes_per_second=313.875Mi/s items_per_second=20.4106M/s null_percent=1 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/10 51924 us 51701 us 13 bytes_per_second=311.891Mi/s items_per_second=20.2816M/s null_percent=10 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/2 51906 us 51690 us 14 bytes_per_second=311.954Mi/s items_per_second=20.2857M/s null_percent=50 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/1 41482 us 41280 us 17 bytes_per_second=390.623Mi/s items_per_second=25.4014M/s null_percent=100 size=1.04858M
SumDoublesGroupedByMediumIntegerSet/1048576/0 50755 us 50578 us 14 bytes_per_second=316.345Mi/s items_per_second=20.732M/s null_percent=0 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/10000 143338 us 142923 us 5 bytes_per_second=258.075Mi/s items_per_second=7.33665M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/100 144967 us 143943 us 5 bytes_per_second=256.246Mi/s items_per_second=7.28467M/s null_percent=1 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/10 144193 us 143649 us 5 bytes_per_second=256.771Mi/s items_per_second=7.29959M/s null_percent=10 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/2 144993 us 144221 us 5 bytes_per_second=255.753Mi/s items_per_second=7.27064M/s null_percent=50 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/1 134475 us 133839 us 5 bytes_per_second=275.59Mi/s items_per_second=7.83458M/s null_percent=100 size=1.04858M
SumDoublesGroupedByTinyIntStringPairSet/1048576/0 143301 us 142909 us 5 bytes_per_second=257.226Mi/s items_per_second=7.33739M/s null_percent=0 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/10000 151412 us 151038 us 4 bytes_per_second=266.496Mi/s items_per_second=6.94245M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/100 151723 us 151318 us 5 bytes_per_second=266.004Mi/s items_per_second=6.92963M/s null_percent=1 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/10 152143 us 151565 us 5 bytes_per_second=265.571Mi/s items_per_second=6.91833M/s null_percent=10 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/2 152183 us 151513 us 5 bytes_per_second=265.661Mi/s items_per_second=6.9207M/s null_percent=50 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/1 141963 us 141269 us 5 bytes_per_second=284.925Mi/s items_per_second=7.42253M/s null_percent=100 size=1.04858M
SumDoublesGroupedBySmallIntStringPairSet/1048576/0 159602 us 153482 us 5 bytes_per_second=261.44Mi/s items_per_second=6.83193M/s null_percent=0 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/10000 156773 us 156236 us 4 bytes_per_second=237.381Mi/s items_per_second=6.7115M/s null_percent=0.01 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/100 156076 us 155671 us 4 bytes_per_second=238.243Mi/s items_per_second=6.73586M/s null_percent=1 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/10 162832 us 159242 us 4 bytes_per_second=232.9Mi/s items_per_second=6.58481M/s null_percent=10 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/2 160253 us 158389 us 4 bytes_per_second=234.155Mi/s items_per_second=6.62027M/s null_percent=50 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/1 151037 us 148625 us 5 bytes_per_second=249.537Mi/s items_per_second=7.05516M/s null_percent=100 size=1.04858M
SumDoublesGroupedByMediumIntStringPairSet/1048576/0 158179 us 156527 us 4 bytes_per_second=236.14Mi/s items_per_second=6.69899M/s null_percent=0 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/10000 58468 us 57843 us 11 bytes_per_second=278.772Mi/s items_per_second=18.128M/s null_percent=0.01 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/100 58101 us 57859 us 12 bytes_per_second=278.696Mi/s items_per_second=18.1231M/s null_percent=1 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/10 58577 us 57867 us 12 bytes_per_second=278.658Mi/s items_per_second=18.1206M/s null_percent=10 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/2 55574 us 55364 us 13 bytes_per_second=291.254Mi/s items_per_second=18.9397M/s null_percent=50 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/1 39821 us 39702 us 18 bytes_per_second=406.146Mi/s items_per_second=26.4109M/s null_percent=100 size=1.04858M
MinMaxDoublesGroupedByMediumInt/1048576/0 56733 us 56608 us 12 bytes_per_second=282.645Mi/s items_per_second=18.5234M/s null_percent=0 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/10000 176193 us 175634 us 4 bytes_per_second=251.295Mi/s items_per_second=5.97025M/s null_percent=0.01 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/100 176881 us 175491 us 4 bytes_per_second=249.705Mi/s items_per_second=5.97508M/s null_percent=1 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/10 165700 us 165194 us 4 bytes_per_second=247.77Mi/s items_per_second=6.34754M/s null_percent=10 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/2 125835 us 124977 us 6 bytes_per_second=225.2Mi/s items_per_second=8.39016M/s null_percent=50 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/1 54546 us 54390 us 13 bytes_per_second=222.928Mi/s items_per_second=19.2789M/s null_percent=100 size=1.04858M
MinMaxShortStringsGroupedByMediumInt/1048576/0 173212 us 172767 us 4 bytes_per_second=254.756Mi/s items_per_second=6.06932M/s null_percent=0 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/10000 173485 us 173082 us 4 bytes_per_second=1.51264Gi/s items_per_second=6.05825M/s null_percent=0.01 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/100 173965 us 173475 us 4 bytes_per_second=1.49513Gi/s items_per_second=6.04455M/s null_percent=1 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/10 165786 us 164827 us 4 bytes_per_second=1.43666Gi/s items_per_second=6.36169M/s null_percent=10 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/2 123678 us 123396 us 5 bytes_per_second=1.10816Gi/s items_per_second=8.49766M/s null_percent=50 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/1 54209 us 54064 us 13 bytes_per_second=224.273Mi/s items_per_second=19.3953M/s null_percent=100 size=1.04858M
MinMaxLongStringsGroupedByMediumInt/1048576/0 177019 us 176798 us 4 bytes_per_second=1.48029Gi/s items_per_second=5.93094M/s null_percent=0 size=1.04858M
SumKernelFloat/1048576/10000 1085 us 1081 us 649 bytes_per_second=924.739Mi/s items_per_second=242.415M/s null_percent=0.01 size=1.04858M
SumKernelFloat/1048576/100 1240 us 1237 us 568 bytes_per_second=808.615Mi/s items_per_second=211.973M/s null_percent=1 size=1.04858M
SumKernelFloat/1048576/10 2229 us 2223 us 314 bytes_per_second=449.777Mi/s items_per_second=117.906M/s null_percent=10 size=1.04858M
SumKernelFloat/1048576/2 4089 us 4078 us 172 bytes_per_second=245.226Mi/s items_per_second=64.2845M/s null_percent=50 size=1.04858M
SumKernelFloat/1048576/1 23.0 us 22.9 us 30624 bytes_per_second=42.607Gi/s items_per_second=11.4372G/s null_percent=100 size=1.04858M
SumKernelFloat/1048576/0 1048 us 1045 us 672 bytes_per_second=956.896Mi/s items_per_second=250.844M/s null_percent=0 size=1.04858M
SumKernelDouble/1048576/10000 533 us 532 us 1297 bytes_per_second=1.83631Gi/s items_per_second=246.465M/s null_percent=0.01 size=1.04858M
SumKernelDouble/1048576/100 621 us 620 us 1134 bytes_per_second=1.57569Gi/s items_per_second=211.486M/s null_percent=1 size=1.04858M
SumKernelDouble/1048576/10 1125 us 1121 us 626 bytes_per_second=891.665Mi/s items_per_second=116.872M/s null_percent=10 size=1.04858M
SumKernelDouble/1048576/2 2051 us 2045 us 341 bytes_per_second=488.925Mi/s items_per_second=64.0843M/s null_percent=50 size=1.04858M
SumKernelDouble/1048576/1 18.0 us 18.0 us 38734 bytes_per_second=54.2857Gi/s items_per_second=7.2861G/s null_percent=100 size=1.04858M
SumKernelDouble/1048576/0 530 us 528 us 1326 bytes_per_second=1.85002Gi/s items_per_second=248.306M/s null_percent=0 size=1.04858M
SumKernelInt8/1048576/10000 2541 us 2514 us 276 bytes_per_second=397.846Mi/s items_per_second=417.172M/s null_percent=0.01 size=1.04858M
SumKernelInt8/1048576/100 3057 us 3015 us 233 bytes_per_second=331.727Mi/s items_per_second=347.841M/s null_percent=1 size=1.04858M
SumKernelInt8/1048576/10 5801 us 5757 us 124 bytes_per_second=173.687Mi/s items_per_second=182.124M/s null_percent=10 size=1.04858M
SumKernelInt8/1048576/2 9913 us 9903 us 71 bytes_per_second=100.98Mi/s items_per_second=105.885M/s null_percent=50 size=1.04858M
SumKernelInt8/1048576/1 132 us 130 us 5431 bytes_per_second=7.49994Gi/s items_per_second=8.053G/s null_percent=100 size=1.04858M
SumKernelInt8/1048576/0 2398 us 2389 us 293 bytes_per_second=418.512Mi/s items_per_second=438.842M/s null_percent=0 size=1.04858M
SumKernelInt16/1048576/10000 1280 us 1272 us 555 bytes_per_second=786.008Mi/s items_per_second=412.094M/s null_percent=0.01 size=1.04858M
SumKernelInt16/1048576/100 1578 us 1529 us 466 bytes_per_second=653.83Mi/s items_per_second=342.795M/s null_percent=1 size=1.04858M
SumKernelInt16/1048576/10 2830 us 2824 us 247 bytes_per_second=354.055Mi/s items_per_second=185.627M/s null_percent=10 size=1.04858M
SumKernelInt16/1048576/2 4983 us 4966 us 141 bytes_per_second=201.35Mi/s items_per_second=105.565M/s null_percent=50 size=1.04858M
SumKernelInt16/1048576/1 71.9 us 71.4 us 9763 bytes_per_second=13.6721Gi/s items_per_second=7.34013G/s null_percent=100 size=1.04858M
SumKernelInt16/1048576/0 1204 us 1201 us 585 bytes_per_second=832.916Mi/s items_per_second=436.688M/s null_percent=0 size=1.04858M
SumKernelInt32/1048576/10000 642 us 640 us 1100 bytes_per_second=1.52524Gi/s items_per_second=409.428M/s null_percent=0.01 size=1.04858M
SumKernelInt32/1048576/100 768 us 765 us 905 bytes_per_second=1.27572Gi/s items_per_second=342.448M/s null_percent=1 size=1.04858M
SumKernelInt32/1048576/10 1431 us 1426 us 488 bytes_per_second=701.039Mi/s items_per_second=183.773M/s null_percent=10 size=1.04858M
SumKernelInt32/1048576/2 2501 us 2492 us 280 bytes_per_second=401.325Mi/s items_per_second=105.205M/s null_percent=50 size=1.04858M
SumKernelInt32/1048576/1 43.4 us 43.2 us 16479 bytes_per_second=22.6253Gi/s items_per_second=6.07344G/s null_percent=100 size=1.04858M
SumKernelInt32/1048576/0 615 us 614 us 1156 bytes_per_second=1.59086Gi/s items_per_second=427.043M/s null_percent=0 size=1.04858M
SumKernelInt64/1048576/10000 330 us 329 us 2105 bytes_per_second=2.96873Gi/s items_per_second=398.456M/s null_percent=0.01 size=1.04858M
SumKernelInt64/1048576/100 391 us 390 us 1797 bytes_per_second=2.50662Gi/s items_per_second=336.433M/s null_percent=1 size=1.04858M
SumKernelInt64/1048576/10 728 us 725 us 961 bytes_per_second=1.34717Gi/s items_per_second=180.814M/s null_percent=10 size=1.04858M
SumKernelInt64/1048576/2 1249 us 1245 us 562 bytes_per_second=802.962Mi/s items_per_second=105.246M/s null_percent=50 size=1.04858M
SumKernelInt64/1048576/1 28.0 us 27.9 us 25043 bytes_per_second=34.9764Gi/s items_per_second=4.69445G/s null_percent=100 size=1.04858M
SumKernelInt64/1048576/0 316 us 314 us 2237 bytes_per_second=3.11055Gi/s items_per_second=417.491M/s null_percent=0 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/10000 2173 us 2161 us 324 bytes_per_second=462.835Mi/s items_per_second=485.317M/s null_percent=0.01 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/100 2179 us 2158 us 319 bytes_per_second=463.407Mi/s items_per_second=485.917M/s null_percent=1 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/10 2204 us 2169 us 323 bytes_per_second=461.041Mi/s items_per_second=483.437M/s null_percent=10 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/2 2157 us 2147 us 327 bytes_per_second=465.685Mi/s items_per_second=488.306M/s null_percent=50 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/1 14.2 us 14.1 us 50201 bytes_per_second=69.4997Gi/s items_per_second=74.6248G/s null_percent=100 size=1.04858M
ModeKernelNarrow<BooleanType>/1048576/0 335 us 333 us 2096 bytes_per_second=2.93092Gi/s items_per_second=3.14705G/s null_percent=0 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/10000 1876 us 1861 us 377 bytes_per_second=537.483Mi/s items_per_second=563.591M/s null_percent=0.01 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/100 2487 us 2359 us 295 bytes_per_second=423.879Mi/s items_per_second=444.47M/s null_percent=1 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/10 5113 us 5096 us 136 bytes_per_second=196.227Mi/s items_per_second=205.759M/s null_percent=10 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/2 9899 us 9844 us 70 bytes_per_second=101.58Mi/s items_per_second=106.515M/s null_percent=50 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/1 18.5 us 18.4 us 38454 bytes_per_second=53.0603Gi/s items_per_second=56.9731G/s null_percent=100 size=1.04858M
ModeKernelNarrow<Int8Type>/1048576/0 1769 us 1752 us 400 bytes_per_second=570.637Mi/s items_per_second=598.356M/s null_percent=0 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/10000 2863 us 2836 us 248 bytes_per_second=352.637Mi/s items_per_second=92.4418M/s null_percent=0.01 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/100 3107 us 3091 us 226 bytes_per_second=323.505Mi/s items_per_second=84.8048M/s null_percent=1 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/10 4245 us 4235 us 164 bytes_per_second=236.116Mi/s items_per_second=61.8965M/s null_percent=10 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/2 6053 us 6024 us 117 bytes_per_second=166.011Mi/s items_per_second=43.5189M/s null_percent=50 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/1 14.3 us 14.3 us 48812 bytes_per_second=68.4243Gi/s items_per_second=18.3675G/s null_percent=100 size=1.04858M
ModeKernelNarrow<Int32Type>/1048576/0 2773 us 2762 us 254 bytes_per_second=362.12Mi/s items_per_second=94.9276M/s null_percent=0 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/10000 1619 us 1612 us 432 bytes_per_second=620.219Mi/s items_per_second=81.2934M/s null_percent=0.01 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/100 1749 us 1742 us 402 bytes_per_second=573.913Mi/s items_per_second=75.2239M/s null_percent=1 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/10 2356 us 2342 us 301 bytes_per_second=427.045Mi/s items_per_second=55.9736M/s null_percent=10 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/2 3215 us 3204 us 218 bytes_per_second=312.099Mi/s items_per_second=40.9075M/s null_percent=50 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/1 14.4 us 14.3 us 48267 bytes_per_second=68.1341Gi/s items_per_second=9.14481G/s null_percent=100 size=1.04858M
ModeKernelNarrow<Int64Type>/1048576/0 1606 us 1588 us 443 bytes_per_second=629.73Mi/s items_per_second=82.54M/s null_percent=0 size=1.04858M
ModeKernelWide<Int32Type>/1048576/10000 25026 us 24923 us 28 bytes_per_second=40.1229Mi/s items_per_second=10.518M/s null_percent=0.01 size=1.04858M
ModeKernelWide<Int32Type>/1048576/100 25576 us 25305 us 28 bytes_per_second=39.5179Mi/s items_per_second=10.3594M/s null_percent=1 size=1.04858M
ModeKernelWide<Int32Type>/1048576/10 25277 us 24883 us 29 bytes_per_second=40.1877Mi/s items_per_second=10.535M/s null_percent=10 size=1.04858M
ModeKernelWide<Int32Type>/1048576/2 17005 us 16871 us 42 bytes_per_second=59.272Mi/s items_per_second=15.5378M/s null_percent=50 size=1.04858M
ModeKernelWide<Int32Type>/1048576/1 14.4 us 14.3 us 48519 bytes_per_second=68.218Gi/s items_per_second=18.3121G/s null_percent=100 size=1.04858M
ModeKernelWide<Int32Type>/1048576/0 25297 us 25144 us 28 bytes_per_second=39.7715Mi/s items_per_second=10.4259M/s null_percent=0 size=1.04858M
ModeKernelWide<Int64Type>/1048576/10000 12311 us 12271 us 58 bytes_per_second=81.4949Mi/s items_per_second=10.6817M/s null_percent=0.01 size=1.04858M
ModeKernelWide<Int64Type>/1048576/100 12286 us 12174 us 57 bytes_per_second=82.1435Mi/s items_per_second=10.7667M/s null_percent=1 size=1.04858M
ModeKernelWide<Int64Type>/1048576/10 11824 us 11794 us 59 bytes_per_second=84.7881Mi/s items_per_second=11.1134M/s null_percent=10 size=1.04858M
ModeKernelWide<Int64Type>/1048576/2 8077 us 8058 us 85 bytes_per_second=124.103Mi/s items_per_second=16.2664M/s null_percent=50 size=1.04858M
ModeKernelWide<Int64Type>/1048576/1 14.6 us 14.5 us 48142 bytes_per_second=67.416Gi/s items_per_second=9.04842G/s null_percent=100 size=1.04858M
ModeKernelWide<Int64Type>/1048576/0 12768 us 12338 us 56 bytes_per_second=81.049Mi/s items_per_second=10.6233M/s null_percent=0 size=1.04858M
ModeKernelWide<FloatType>/1048576/10000 28289 us 28176 us 25 bytes_per_second=35.4915Mi/s items_per_second=9.30387M/s null_percent=0.01 size=1.04858M
ModeKernelWide<FloatType>/1048576/100 31624 us 28704 us 25 bytes_per_second=34.8386Mi/s items_per_second=9.13274M/s null_percent=1 size=1.04858M
ModeKernelWide<FloatType>/1048576/10 29644 us 27275 us 26 bytes_per_second=36.664Mi/s items_per_second=9.61124M/s null_percent=10 size=1.04858M
ModeKernelWide<FloatType>/1048576/2 16866 us 16448 us 43 bytes_per_second=60.7964Mi/s items_per_second=15.9374M/s null_percent=50 size=1.04858M
ModeKernelWide<FloatType>/1048576/1 15.0 us 14.8 us 44682 bytes_per_second=65.8006Gi/s items_per_second=17.6632G/s null_percent=100 size=1.04858M
ModeKernelWide<FloatType>/1048576/0 30706 us 29282 us 24 bytes_per_second=34.1505Mi/s items_per_second=8.95234M/s null_percent=0 size=1.04858M
ModeKernelWide<DoubleType>/1048576/10000 15121 us 14549 us 48 bytes_per_second=68.7349Mi/s items_per_second=9.00922M/s null_percent=0.01 size=1.04858M
ModeKernelWide<DoubleType>/1048576/100 13874 us 13835 us 49 bytes_per_second=72.28Mi/s items_per_second=9.47389M/s null_percent=1 size=1.04858M
ModeKernelWide<DoubleType>/1048576/10 13105 us 12975 us 55 bytes_per_second=77.0726Mi/s items_per_second=10.1021M/s null_percent=10 size=1.04858M
ModeKernelWide<DoubleType>/1048576/2 9270 us 8607 us 85 bytes_per_second=116.185Mi/s items_per_second=15.2286M/s null_percent=50 size=1.04858M
ModeKernelWide<DoubleType>/1048576/1 14.8 us 14.7 us 48444 bytes_per_second=66.5142Gi/s items_per_second=8.92738G/s null_percent=100 size=1.04858M
ModeKernelWide<DoubleType>/1048576/0 14152 us 14078 us 50 bytes_per_second=71.0308Mi/s items_per_second=9.31015M/s null_percent=0 size=1.04858M
MinMaxKernelFloat/1048576/10000 2022 us 2011 us 348 bytes_per_second=497.327Mi/s items_per_second=130.371M/s null_percent=0.01 size=1.04858M
MinMaxKernelFloat/1048576/100 2567 us 2556 us 274 bytes_per_second=391.168Mi/s items_per_second=102.542M/s null_percent=1 size=1.04858M
MinMaxKernelFloat/1048576/10 3152 us 3136 us 226 bytes_per_second=318.917Mi/s items_per_second=83.6022M/s null_percent=10 size=1.04858M
MinMaxKernelFloat/1048576/2 3462 us 3449 us 203 bytes_per_second=289.918Mi/s items_per_second=76.0004M/s null_percent=50 size=1.04858M
MinMaxKernelFloat/1048576/1 79.9 us 79.5 us 8887 bytes_per_second=12.2816Gi/s items_per_second=3.29681G/s null_percent=100 size=1.04858M
MinMaxKernelFloat/1048576/0 2369 us 2363 us 296 bytes_per_second=423.141Mi/s items_per_second=110.924M/s null_percent=0 size=1.04858M
MinMaxKernelDouble/1048576/10000 797 us 789 us 895 bytes_per_second=1.23724Gi/s items_per_second=166.059M/s null_percent=0.01 size=1.04858M
MinMaxKernelDouble/1048576/100 1065 us 1062 us 658 bytes_per_second=941.807Mi/s items_per_second=123.445M/s null_percent=1 size=1.04858M
MinMaxKernelDouble/1048576/10 1377 us 1371 us 512 bytes_per_second=729.653Mi/s items_per_second=95.6371M/s null_percent=10 size=1.04858M
MinMaxKernelDouble/1048576/2 1609 us 1596 us 440 bytes_per_second=626.459Mi/s items_per_second=82.1113M/s null_percent=50 size=1.04858M
MinMaxKernelDouble/1048576/1 55.0 us 54.2 us 13132 bytes_per_second=18.0271Gi/s items_per_second=2.41955G/s null_percent=100 size=1.04858M
MinMaxKernelDouble/1048576/0 1193 us 1187 us 586 bytes_per_second=842.442Mi/s items_per_second=110.421M/s null_percent=0 size=1.04858M
MinMaxKernelInt8/1048576/10000 11490 us 11374 us 61 bytes_per_second=87.9218Mi/s items_per_second=92.1927M/s null_percent=0.01 size=1.04858M
MinMaxKernelInt8/1048576/100 13860 us 13715 us 50 bytes_per_second=72.912Mi/s items_per_second=76.4538M/s null_percent=1 size=1.04858M
MinMaxKernelInt8/1048576/10 16427 us 15897 us 45 bytes_per_second=62.9057Mi/s items_per_second=65.9614M/s null_percent=10 size=1.04858M
MinMaxKernelInt8/1048576/2 16097 us 15612 us 45 bytes_per_second=64.0518Mi/s items_per_second=67.1632M/s null_percent=50 size=1.04858M
MinMaxKernelInt8/1048576/1 234 us 233 us 3012 bytes_per_second=4.19537Gi/s items_per_second=4.50474G/s null_percent=100 size=1.04858M
MinMaxKernelInt8/1048576/0 13340 us 13279 us 53 bytes_per_second=75.3054Mi/s items_per_second=78.9634M/s null_percent=0 size=1.04858M
MinMaxKernelInt16/1048576/10000 5671 us 5641 us 125 bytes_per_second=177.27Mi/s items_per_second=92.9405M/s null_percent=0.01 size=1.04858M
MinMaxKernelInt16/1048576/100 7012 us 6932 us 94 bytes_per_second=144.268Mi/s items_per_second=75.6381M/s null_percent=1 size=1.04858M
MinMaxKernelInt16/1048576/10 8389 us 8018 us 88 bytes_per_second=124.714Mi/s items_per_second=65.386M/s null_percent=10 size=1.04858M
MinMaxKernelInt16/1048576/2 8013 us 7946 us 87 bytes_per_second=125.853Mi/s items_per_second=65.9835M/s null_percent=50 size=1.04858M
MinMaxKernelInt16/1048576/1 138 us 134 us 5210 bytes_per_second=7.30678Gi/s items_per_second=3.9228G/s null_percent=100 size=1.04858M
MinMaxKernelInt16/1048576/0 7026 us 6802 us 104 bytes_per_second=147.024Mi/s items_per_second=77.0829M/s null_percent=0 size=1.04858M
MinMaxKernelInt32/1048576/10000 2865 us 2836 us 245 bytes_per_second=352.665Mi/s items_per_second=92.4489M/s null_percent=0.01 size=1.04858M
MinMaxKernelInt32/1048576/100 3438 us 3419 us 204 bytes_per_second=292.444Mi/s items_per_second=76.6624M/s null_percent=1 size=1.04858M
MinMaxKernelInt32/1048576/10 3959 us 3937 us 178 bytes_per_second=254.013Mi/s items_per_second=66.5879M/s null_percent=10 size=1.04858M
MinMaxKernelInt32/1048576/2 3909 us 3901 us 179 bytes_per_second=256.324Mi/s items_per_second=67.1938M/s null_percent=50 size=1.04858M
MinMaxKernelInt32/1048576/1 79.4 us 79.0 us 8806 bytes_per_second=12.354Gi/s items_per_second=3.31626G/s null_percent=100 size=1.04858M
MinMaxKernelInt32/1048576/0 3387 us 3366 us 207 bytes_per_second=297.085Mi/s items_per_second=77.8791M/s null_percent=0 size=1.04858M
MinMaxKernelInt64/1048576/10000 1651 us 1625 us 430 bytes_per_second=615.379Mi/s items_per_second=80.659M/s null_percent=0.01 size=1.04858M
MinMaxKernelInt64/1048576/100 1883 us 1878 us 371 bytes_per_second=532.517Mi/s items_per_second=69.798M/s null_percent=1 size=1.04858M
MinMaxKernelInt64/1048576/10 2132 us 2127 us 327 bytes_per_second=470.096Mi/s items_per_second=61.6165M/s null_percent=10 size=1.04858M
MinMaxKernelInt64/1048576/2 1997 us 1990 us 353 bytes_per_second=502.578Mi/s items_per_second=65.8739M/s null_percent=50 size=1.04858M
MinMaxKernelInt64/1048576/1 53.4 us 53.3 us 13130 bytes_per_second=18.3342Gi/s items_per_second=2.46077G/s null_percent=100 size=1.04858M
MinMaxKernelInt64/1048576/0 1850 us 1844 us 379 bytes_per_second=542.408Mi/s items_per_second=71.0945M/s null_percent=0 size=1.04858M
CountKernelBenchInt64/1048576/2 16289 ns 16255 ns 43117 bytes_per_second=60.0777Gi/s items_per_second=8.06349G/s null_percent=50 size=1.04858M
VarianceKernelInt32/1048576/10000 868 us 867 us 809 bytes_per_second=1.12696Gi/s items_per_second=302.516M/s null_percent=0.01 size=1.04858M
VarianceKernelInt32/1048576/100 995 us 992 us 702 bytes_per_second=1008.44Mi/s items_per_second=264.357M/s null_percent=1 size=1.04858M
VarianceKernelInt32/1048576/10 1612 us 1608 us 435 bytes_per_second=621.816Mi/s items_per_second=163.005M/s null_percent=10 size=1.04858M
VarianceKernelInt32/1048576/2 2605 us 2599 us 270 bytes_per_second=384.743Mi/s items_per_second=100.858M/s null_percent=50 size=1.04858M
VarianceKernelInt32/1048576/1 21.8 us 21.8 us 32223 bytes_per_second=44.8025Gi/s items_per_second=12.0266G/s null_percent=100 size=1.04858M
VarianceKernelInt32/1048576/0 812 us 810 us 865 bytes_per_second=1.2062Gi/s items_per_second=323.787M/s null_percent=0 size=1.04858M
VarianceKernelInt64/1048576/10000 1068 us 1066 us 656 bytes_per_second=938.183Mi/s items_per_second=122.97M/s null_percent=0.01 size=1.04858M
VarianceKernelInt64/1048576/100 1249 us 1246 us 567 bytes_per_second=802.413Mi/s items_per_second=105.174M/s null_percent=1 size=1.04858M
VarianceKernelInt64/1048576/10 2006 us 1997 us 349 bytes_per_second=500.847Mi/s items_per_second=65.647M/s null_percent=10 size=1.04858M
VarianceKernelInt64/1048576/2 3415 us 3408 us 205 bytes_per_second=293.46Mi/s items_per_second=38.4644M/s null_percent=50 size=1.04858M
VarianceKernelInt64/1048576/1 16.8 us 16.8 us 41537 bytes_per_second=58.0886Gi/s items_per_second=7.79652G/s null_percent=100 size=1.04858M
VarianceKernelInt64/1048576/0 1050 us 1047 us 671 bytes_per_second=954.803Mi/s items_per_second=125.148M/s null_percent=0 size=1.04858M
VarianceKernelFloat/1048576/10000 2529 us 2523 us 275 bytes_per_second=396.344Mi/s items_per_second=103.899M/s null_percent=0.01 size=1.04858M
VarianceKernelFloat/1048576/100 2897 us 2890 us 242 bytes_per_second=346.016Mi/s items_per_second=90.706M/s null_percent=1 size=1.04858M
VarianceKernelFloat/1048576/10 4791 us 4781 us 147 bytes_per_second=209.173Mi/s items_per_second=54.8335M/s null_percent=10 size=1.04858M
VarianceKernelFloat/1048576/2 8444 us 8427 us 83 bytes_per_second=118.662Mi/s items_per_second=31.1066M/s null_percent=50 size=1.04858M
VarianceKernelFloat/1048576/1 21.7 us 21.6 us 32353 bytes_per_second=45.1365Gi/s items_per_second=12.1163G/s null_percent=100 size=1.04858M
VarianceKernelFloat/1048576/0 2506 us 2500 us 278 bytes_per_second=399.983Mi/s items_per_second=104.853M/s null_percent=0 size=1.04858M
VarianceKernelDouble/1048576/10000 1268 us 1267 us 552 bytes_per_second=789.199Mi/s items_per_second=103.442M/s null_percent=0.01 size=1.04858M
VarianceKernelDouble/1048576/100 1454 us 1451 us 483 bytes_per_second=689.415Mi/s items_per_second=90.363M/s null_percent=1 size=1.04858M
VarianceKernelDouble/1048576/10 2394 us 2388 us 292 bytes_per_second=418.68Mi/s items_per_second=54.8772M/s null_percent=10 size=1.04858M
VarianceKernelDouble/1048576/2 4244 us 4236 us 165 bytes_per_second=236.089Mi/s items_per_second=30.9446M/s null_percent=50 size=1.04858M
VarianceKernelDouble/1048576/1 16.8 us 16.7 us 41958 bytes_per_second=58.3403Gi/s items_per_second=7.8303G/s null_percent=100 size=1.04858M
VarianceKernelDouble/1048576/0 1272 us 1270 us 559 bytes_per_second=787.508Mi/s items_per_second=103.22M/s null_percent=0 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/10000 3319 us 3295 us 213 bytes_per_second=303.533Mi/s items_per_second=79.5695M/s null_percent=0.01 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/100 3552 us 3545 us 198 bytes_per_second=282.071Mi/s items_per_second=73.9432M/s null_percent=1 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/10 4704 us 4695 us 149 bytes_per_second=213Mi/s items_per_second=55.8368M/s null_percent=10 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/2 6410 us 6365 us 111 bytes_per_second=157.117Mi/s items_per_second=41.1872M/s null_percent=50 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/1 18.3 us 18.3 us 37897 bytes_per_second=53.3437Gi/s items_per_second=14.3193G/s null_percent=100 size=1.04858M
QuantileKernelMedianNarrow<Int32Type>/1048576/0 3218 us 3213 us 218 bytes_per_second=311.246Mi/s items_per_second=81.5912M/s null_percent=0 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/10000 13096 us 13069 us 53 bytes_per_second=76.5156Mi/s items_per_second=20.0581M/s null_percent=0.01 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/100 11881 us 11853 us 59 bytes_per_second=84.3674Mi/s items_per_second=22.1164M/s null_percent=1 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/10 13529 us 13500 us 52 bytes_per_second=74.0744Mi/s items_per_second=19.4182M/s null_percent=10 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/2 11157 us 11132 us 63 bytes_per_second=89.8348Mi/s items_per_second=23.5497M/s null_percent=50 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/1 18.6 us 18.5 us 38261 bytes_per_second=52.8188Gi/s items_per_second=14.1784G/s null_percent=100 size=1.04858M
QuantileKernelMedianWide<Int32Type>/1048576/0 13340 us 13288 us 53 bytes_per_second=75.2575Mi/s items_per_second=19.7283M/s null_percent=0 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/10000 2037 us 2030 us 344 bytes_per_second=492.663Mi/s items_per_second=64.5743M/s null_percent=0.01 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/100 2171 us 2163 us 325 bytes_per_second=462.413Mi/s items_per_second=60.6094M/s null_percent=1 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/10 2819 us 2793 us 246 bytes_per_second=357.985Mi/s items_per_second=46.9218M/s null_percent=10 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/2 3873 us 3817 us 183 bytes_per_second=261.982Mi/s items_per_second=34.3386M/s null_percent=50 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/1 18.7 us 18.6 us 37396 bytes_per_second=52.5835Gi/s items_per_second=7.05763G/s null_percent=100 size=1.04858M
QuantileKernelMedianNarrow<Int64Type>/1048576/0 2013 us 2003 us 347 bytes_per_second=499.212Mi/s items_per_second=65.4327M/s null_percent=0 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/10000 7798 us 7642 us 94 bytes_per_second=130.853Mi/s items_per_second=17.1512M/s null_percent=0.01 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/100 7871 us 7784 us 90 bytes_per_second=128.475Mi/s items_per_second=16.8394M/s null_percent=1 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/10 7299 us 6993 us 100 bytes_per_second=143.004Mi/s items_per_second=18.7438M/s null_percent=10 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/2 4318 us 4263 us 165 bytes_per_second=234.569Mi/s items_per_second=30.7455M/s null_percent=50 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/1 20.0 us 19.3 us 35730 bytes_per_second=50.5007Gi/s items_per_second=6.77809G/s null_percent=100 size=1.04858M
QuantileKernelMedianWide<Int64Type>/1048576/0 6949 us 6905 us 101 bytes_per_second=144.821Mi/s items_per_second=18.982M/s null_percent=0 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/10000 5727 us 5696 us 121 bytes_per_second=175.576Mi/s items_per_second=23.0131M/s null_percent=0.01 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/100 7514 us 7461 us 92 bytes_per_second=134.028Mi/s items_per_second=17.5673M/s null_percent=1 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/10 6402 us 6367 us 109 bytes_per_second=157.066Mi/s items_per_second=20.5869M/s null_percent=10 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/2 4149 us 4131 us 170 bytes_per_second=242.07Mi/s items_per_second=31.7286M/s null_percent=50 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/1 18.6 us 18.5 us 37914 bytes_per_second=52.7784Gi/s items_per_second=7.08379G/s null_percent=100 size=1.04858M
QuantileKernelMedianWide<DoubleType>/1048576/0 6983 us 6960 us 100 bytes_per_second=143.67Mi/s items_per_second=18.8312M/s null_percent=0 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/10000 3402 us 3392 us 205 bytes_per_second=294.836Mi/s items_per_second=77.2895M/s null_percent=0.01 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/100 3662 us 3651 us 187 bytes_per_second=273.918Mi/s items_per_second=71.8059M/s null_percent=1 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/10 4867 us 4843 us 142 bytes_per_second=206.487Mi/s items_per_second=54.1295M/s null_percent=10 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/2 6646 us 6593 us 108 bytes_per_second=151.665Mi/s items_per_second=39.7579M/s null_percent=50 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/1 18.7 us 18.6 us 37248 bytes_per_second=52.4123Gi/s items_per_second=14.0693G/s null_percent=100 size=1.04858M
QuantileKernelDecilesNarrow<Int32Type>/1048576/0 3427 us 3387 us 207 bytes_per_second=295.282Mi/s items_per_second=77.4064M/s null_percent=0 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/10000 30700 us 30029 us 23 bytes_per_second=33.3012Mi/s items_per_second=8.72972M/s null_percent=0.01 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/100 28767 us 28602 us 24 bytes_per_second=34.9621Mi/s items_per_second=9.1651M/s null_percent=1 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/10 28405 us 28260 us 25 bytes_per_second=35.3863Mi/s items_per_second=9.27631M/s null_percent=10 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/2 19939 us 19814 us 35 bytes_per_second=50.4683Mi/s items_per_second=13.23M/s null_percent=50 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/1 18.9 us 18.8 us 37463 bytes_per_second=51.9337Gi/s items_per_second=13.9408G/s null_percent=100 size=1.04858M
QuantileKernelDecilesWide<Int32Type>/1048576/0 29720 us 29610 us 24 bytes_per_second=33.7729Mi/s items_per_second=8.85337M/s null_percent=0 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/10000 15470 us 15396 us 45 bytes_per_second=64.9502Mi/s items_per_second=8.51316M/s null_percent=0.01 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/100 14742 us 14709 us 47 bytes_per_second=67.9869Mi/s items_per_second=8.91117M/s null_percent=1 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/10 14383 us 14163 us 50 bytes_per_second=70.6081Mi/s items_per_second=9.25475M/s null_percent=10 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/2 8639 us 8585 us 81 bytes_per_second=116.477Mi/s items_per_second=15.2669M/s null_percent=50 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/1 19.1 us 18.8 us 36799 bytes_per_second=51.8316Gi/s items_per_second=6.95672G/s null_percent=100 size=1.04858M
QuantileKernelDecilesWide<DoubleType>/1048576/0 15781 us 15748 us 44 bytes_per_second=63.5003Mi/s items_per_second=8.32311M/s null_percent=0 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/10000 3401 us 3395 us 201 bytes_per_second=294.521Mi/s items_per_second=77.2069M/s null_percent=0.01 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/100 3728 us 3695 us 191 bytes_per_second=270.633Mi/s items_per_second=70.9449M/s null_percent=1 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/10 5022 us 4934 us 141 bytes_per_second=202.688Mi/s items_per_second=53.1335M/s null_percent=10 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/2 6745 us 6651 us 106 bytes_per_second=150.361Mi/s items_per_second=39.4163M/s null_percent=50 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/1 20.5 us 20.2 us 34459 bytes_per_second=48.3107Gi/s items_per_second=12.9683G/s null_percent=100 size=1.04858M
QuantileKernelCentilesNarrow<Int32Type>/1048576/0 3404 us 3372 us 206 bytes_per_second=296.593Mi/s items_per_second=77.75M/s null_percent=0 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/10000 111628 us 110798 us 6 bytes_per_second=9.02545Mi/s items_per_second=2.36597M/s null_percent=0.01 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/100 112297 us 111185 us 6 bytes_per_second=8.99402Mi/s items_per_second=2.35773M/s null_percent=1 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/10 105556 us 104456 us 7 bytes_per_second=9.57342Mi/s items_per_second=2.50962M/s null_percent=10 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/2 61667 us 61033 us 12 bytes_per_second=16.3846Mi/s items_per_second=4.29511M/s null_percent=50 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/1 20.2 us 20.0 us 34946 bytes_per_second=48.7918Gi/s items_per_second=13.0975G/s null_percent=100 size=1.04858M
QuantileKernelCentilesWide<Int32Type>/1048576/0 114918 us 114062 us 6 bytes_per_second=8.76714Mi/s items_per_second=2.29825M/s null_percent=0 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/10000 57672 us 57335 us 12 bytes_per_second=17.4415Mi/s items_per_second=2.28609M/s null_percent=0.01 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/100 56663 us 56301 us 12 bytes_per_second=17.7618Mi/s items_per_second=2.32808M/s null_percent=1 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/10 53007 us 52624 us 13 bytes_per_second=19.0027Mi/s items_per_second=2.49072M/s null_percent=10 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/2 29992 us 29779 us 23 bytes_per_second=33.5805Mi/s items_per_second=4.40147M/s null_percent=50 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/1 20.7 us 20.5 us 34404 bytes_per_second=47.7271Gi/s items_per_second=6.40583G/s null_percent=100 size=1.04858M
QuantileKernelCentilesWide<DoubleType>/1048576/0 58754 us 58149 us 12 bytes_per_second=17.1972Mi/s items_per_second=2.25407M/s null_percent=0 size=1.04858M
TDigestKernelDoubleMedian/1048576/10000 10611 us 10536 us 66 bytes_per_second=94.9169Mi/s items_per_second=12.4409M/s null_percent=0.01 size=1.04858M
TDigestKernelDoubleMedian/1048576/100 10461 us 10419 us 67 bytes_per_second=95.9818Mi/s items_per_second=12.5805M/s null_percent=1 size=1.04858M
TDigestKernelDoubleMedian/1048576/10 10113 us 9991 us 70 bytes_per_second=100.086Mi/s items_per_second=13.1184M/s null_percent=10 size=1.04858M
TDigestKernelDoubleMedian/1048576/2 6490 us 6461 us 105 bytes_per_second=154.774Mi/s items_per_second=20.2865M/s null_percent=50 size=1.04858M
TDigestKernelDoubleMedian/1048576/1 27.9 us 27.8 us 25181 bytes_per_second=35.1285Gi/s items_per_second=4.71487G/s null_percent=100 size=1.04858M
TDigestKernelDoubleMedian/1048576/0 10439 us 10401 us 67 bytes_per_second=96.1409Mi/s items_per_second=12.6014M/s null_percent=0 size=1.04858M
TDigestKernelDoubleDeciles/1048576/10000 10492 us 10444 us 67 bytes_per_second=95.752Mi/s items_per_second=12.5504M/s null_percent=0.01 size=1.04858M
TDigestKernelDoubleDeciles/1048576/100 10457 us 10417 us 67 bytes_per_second=96.0009Mi/s items_per_second=12.583M/s null_percent=1 size=1.04858M
TDigestKernelDoubleDeciles/1048576/10 9996 us 9947 us 70 bytes_per_second=100.534Mi/s items_per_second=13.1772M/s null_percent=10 size=1.04858M
TDigestKernelDoubleDeciles/1048576/2 6635 us 6538 us 107 bytes_per_second=152.941Mi/s items_per_second=20.0463M/s null_percent=50 size=1.04858M
TDigestKernelDoubleDeciles/1048576/1 28.5 us 28.3 us 24561 bytes_per_second=34.5396Gi/s items_per_second=4.63583G/s null_percent=100 size=1.04858M
TDigestKernelDoubleDeciles/1048576/0 10636 us 10540 us 66 bytes_per_second=94.8757Mi/s items_per_second=12.4355M/s null_percent=0 size=1.04858M
TDigestKernelDoubleCentiles/1048576/10000 10721 us 10595 us 66 bytes_per_second=94.3825Mi/s items_per_second=12.3709M/s null_percent=0.01 size=1.04858M
TDigestKernelDoubleCentiles/1048576/100 10991 us 10743 us 65 bytes_per_second=93.0823Mi/s items_per_second=12.2005M/s null_percent=1 size=1.04858M
TDigestKernelDoubleCentiles/1048576/10 10213 us 10090 us 69 bytes_per_second=99.1086Mi/s items_per_second=12.9904M/s null_percent=10 size=1.04858M
TDigestKernelDoubleCentiles/1048576/2 6573 us 6525 us 107 bytes_per_second=153.268Mi/s items_per_second=20.0891M/s null_percent=50 size=1.04858M
TDigestKernelDoubleCentiles/1048576/1 30.1 us 29.7 us 23694 bytes_per_second=32.8311Gi/s items_per_second=4.40652G/s null_percent=100 size=1.04858M
TDigestKernelDoubleCentiles/1048576/0 10653 us 10597 us 65 bytes_per_second=94.3707Mi/s items_per_second=12.3694M/s null_percent=0 size=1.04858M
```
</details>
#### asof-join
<details><summary>result</summary>
```
Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
***WARNING*** Failed to set thread affinity. Estimated CPU frequency may be incorrect.
2024-03-29T19:04:59+09:00
Running /Users/lama/workspace/arrow-new/cpp/cmake-build-debug/debug/arrow-acero-asof-join-benchmark
Run on (10 X 24 MHz CPU s)
CPU Caches:
L1 Data 64 KiB
L1 Instruction 128 KiB
L2 Unified 4096 KiB (x10)
Load Average: 6.94, 15.06, 15.61
***WARNING*** Library was built as DEBUG. Timings may be affected.
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations UserCounters...
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
AsOfJoinOverhead/left_freq:200/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:200/right_cols:20/right_ids:500/real_time 628210458 ns 31379000 ns 1 bytes_per_second=44.0808M/s rows_per_second=256.284k/s maximum_peak_memory=29.0547M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 325281625 ns 12674500 ns 2 bytes_per_second=42.8306M/s rows_per_second=249.015k/s maximum_peak_memory=29.0547M
AsOfJoinOverhead/left_freq:1000/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:1000/right_cols:20/right_ids:500/real_time 127967508 ns 5837000 ns 5 bytes_per_second=44.355M/s rows_per_second=257.878k/s maximum_peak_memory=29.0547M
AsOfJoinOverhead/left_freq:400/left_cols:10/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:10/right_ids:500/real_time 218999362 ns 6874667 ns 3 bytes_per_second=34.0275M/s rows_per_second=369.864k/s maximum_peak_memory=29.0547M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 311665813 ns 12404500 ns 2 bytes_per_second=44.7017M/s rows_per_second=259.894k/s maximum_peak_memory=29.0547M
AsOfJoinOverhead/left_freq:400/left_cols:100/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:100/right_ids:500/real_time 974169334 ns 56985000 ns 1 bytes_per_second=67.516M/s rows_per_second=83.1478k/s maximum_peak_memory=72.2598M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:100/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:100/real_time 62534064 ns 3492000 ns 11 bytes_per_second=44.5581M/s rows_per_second=259.059k/s maximum_peak_memory=72.2598M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 308078958 ns 12186500 ns 2 bytes_per_second=45.2222M/s rows_per_second=262.92k/s maximum_peak_memory=72.2598M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:1000/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:1000/real_time 620623042 ns 23515000 ns 1 bytes_per_second=44.8968M/s rows_per_second=261.028k/s maximum_peak_memory=72.2598M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 313503645 ns 12187500 ns 2 bytes_per_second=44.4397M/s rows_per_second=258.37k/s maximum_peak_memory=72.2598M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:10/right_freq:400/right_cols:20/right_ids:500/real_time 1836709125 ns 66782000 ns 1 bytes_per_second=41.7192M/s rows_per_second=242.553k/s maximum_peak_memory=83.9027M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:50/right_freq:400/right_cols:20/right_ids:500/real_time 8678814916 ns 307575000 ns 1 bytes_per_second=40.9349M/s rows_per_second=237.993k/s maximum_peak_memory=388.827M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:1000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 354310062 ns 44033500 ns 2 bytes_per_second=39.3215M/s rows_per_second=228.613k/s maximum_peak_memory=388.827M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:4000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 302512729 ns 11968000 ns 2 bytes_per_second=46.0543M/s rows_per_second=267.757k/s maximum_peak_memory=388.827M
AsOfJoinOverhead/left_freq:400/left_cols:20/left_ids:500/batch_size:32000/num_right_tables:1/right_freq:400/right_cols:20/right_ids:500/real_time 303414416 ns 2661000 ns 2 bytes_per_second=45.9174M/s rows_per_second=266.962k/s maximum_peak_memory=388.827M
```
</details>
#### expression
<details><summary>result</summary>
```
Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
***WARNING*** Failed to set thread affinity. Estimated CPU frequency may be incorrect.
2024-03-29T18:57:43+09:00
Running /Users/lama/workspace/arrow-new/cpp/cmake-build-debug/debug/arrow-acero-expression-benchmark
Run on (10 X 24 MHz CPU s)
CPU Caches:
L1 Data 64 KiB
L1 Instruction 128 KiB
L2 Unified 4096 KiB (x10)
Load Average: 28.69, 20.46, 16.01
***WARNING*** Library was built as DEBUG. Timings may be affected.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations UserCounters...
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:1000/real_time/threads:1 77697 ns 77502 ns 8867 batches_per_second=12.8706M/s rows_per_second=12.8706G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:1000/real_time/threads:10 10884 ns 104100 ns 64740 batches_per_second=91.8818M/s rows_per_second=91.8818G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:10000/real_time/threads:1 77359 ns 77200 ns 9025 batches_per_second=1.29267M/s rows_per_second=12.9267G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:10000/real_time/threads:10 11776 ns 100244 ns 64090 batches_per_second=8.4921M/s rows_per_second=84.921G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:100000/real_time/threads:1 79272 ns 78743 ns 8735 batches_per_second=126.147k/s rows_per_second=12.6147G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:100000/real_time/threads:10 11191 ns 101687 ns 62390 batches_per_second=893.567k/s rows_per_second=89.3567G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:1000000/real_time/threads:1 78352 ns 78082 ns 9003 batches_per_second=12.7629k/s rows_per_second=12.7629G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_simple/rows_per_batch:1000000/real_time/threads:10 11901 ns 100328 ns 58990 batches_per_second=84.0242k/s rows_per_second=84.0242G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:1000/real_time/threads:1 119411 ns 119065 ns 5888 batches_per_second=8.37447M/s rows_per_second=8.37447G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:1000/real_time/threads:10 16480 ns 154154 ns 41600 batches_per_second=60.6804M/s rows_per_second=60.6804G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:10000/real_time/threads:1 120488 ns 119257 ns 5862 batches_per_second=829.959k/s rows_per_second=8.29959G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:10000/real_time/threads:10 16375 ns 154219 ns 42360 batches_per_second=6.10686M/s rows_per_second=61.0686G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:100000/real_time/threads:1 120666 ns 119012 ns 5839 batches_per_second=82.8736k/s rows_per_second=8.28736G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:100000/real_time/threads:10 17693 ns 155204 ns 43000 batches_per_second=565.179k/s rows_per_second=56.5179G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:1000000/real_time/threads:1 120706 ns 119156 ns 5832 batches_per_second=8.28458k/s rows_per_second=8.28458G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_simple/rows_per_batch:1000000/real_time/threads:10 16913 ns 154222 ns 42870 batches_per_second=59.1246k/s rows_per_second=59.1246G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:1000/real_time/threads:1 173835 ns 171933 ns 4057 batches_per_second=5.75258M/s rows_per_second=5.75258G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:1000/real_time/threads:10 24345 ns 221651 ns 27050 batches_per_second=41.0762M/s rows_per_second=41.0762G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:10000/real_time/threads:1 173772 ns 171772 ns 4011 batches_per_second=575.466k/s rows_per_second=5.75466G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:10000/real_time/threads:10 24807 ns 218394 ns 26510 batches_per_second=4.03106M/s rows_per_second=40.3106G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:100000/real_time/threads:1 173191 ns 171652 ns 4045 batches_per_second=57.7397k/s rows_per_second=5.77397G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:100000/real_time/threads:10 25449 ns 216748 ns 28020 batches_per_second=392.937k/s rows_per_second=39.2937G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:1 171021 ns 168744 ns 4078 batches_per_second=5.84725k/s rows_per_second=5.84725G/s
SimplifyFilterWithGuarantee/negative_filter_simple_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:10 23786 ns 216930 ns 27440 batches_per_second=42.0412k/s rows_per_second=42.0412G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:1000/real_time/threads:1 213321 ns 210503 ns 3275 batches_per_second=4.68776M/s rows_per_second=4.68776G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:1000/real_time/threads:10 30962 ns 271477 ns 23390 batches_per_second=32.2978M/s rows_per_second=32.2978G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:10000/real_time/threads:1 215038 ns 212574 ns 3251 batches_per_second=465.035k/s rows_per_second=4.65035G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:10000/real_time/threads:10 32866 ns 268400 ns 23300 batches_per_second=3.04266M/s rows_per_second=30.4266G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:100000/real_time/threads:1 234130 ns 223859 ns 3259 batches_per_second=42.7114k/s rows_per_second=4.27114G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:100000/real_time/threads:10 45315 ns 276181 ns 19710 batches_per_second=220.677k/s rows_per_second=22.0677G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:1 221907 ns 217273 ns 3054 batches_per_second=4.50638k/s rows_per_second=4.50638G/s
SimplifyFilterWithGuarantee/negative_filter_cast_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:10 30219 ns 269786 ns 19960 batches_per_second=33.0915k/s rows_per_second=33.0915G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:1000/real_time/threads:1 78455 ns 77530 ns 8731 batches_per_second=12.7462M/s rows_per_second=12.7462G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:1000/real_time/threads:10 11538 ns 101031 ns 63740 batches_per_second=86.6675M/s rows_per_second=86.6675G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:10000/real_time/threads:1 79036 ns 77964 ns 8927 batches_per_second=1.26525M/s rows_per_second=12.6525G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:10000/real_time/threads:10 11504 ns 99717 ns 62810 batches_per_second=8.69279M/s rows_per_second=86.9279G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:100000/real_time/threads:1 79713 ns 78485 ns 8768 batches_per_second=125.449k/s rows_per_second=12.5449G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:100000/real_time/threads:10 11129 ns 99681 ns 61790 batches_per_second=898.531k/s rows_per_second=89.8531G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:1000000/real_time/threads:1 78720 ns 77706 ns 8909 batches_per_second=12.7033k/s rows_per_second=12.7033G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_simple/rows_per_batch:1000000/real_time/threads:10 11658 ns 99319 ns 63100 batches_per_second=85.7811k/s rows_per_second=85.7811G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:1000/real_time/threads:1 125164 ns 122406 ns 5711 batches_per_second=7.98953M/s rows_per_second=7.98953G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:1000/real_time/threads:10 18044 ns 151094 ns 41920 batches_per_second=55.4193M/s rows_per_second=55.4193G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:10000/real_time/threads:1 123256 ns 120933 ns 5475 batches_per_second=811.321k/s rows_per_second=8.11321G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:10000/real_time/threads:10 16910 ns 146805 ns 42220 batches_per_second=5.91361M/s rows_per_second=59.1361G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:100000/real_time/threads:1 121570 ns 120000 ns 5773 batches_per_second=82.2574k/s rows_per_second=8.22574G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:100000/real_time/threads:10 16947 ns 148156 ns 40880 batches_per_second=590.066k/s rows_per_second=59.0066G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:1000000/real_time/threads:1 129572 ns 123232 ns 5272 batches_per_second=7.71774k/s rows_per_second=7.71774G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_simple/rows_per_batch:1000000/real_time/threads:10 17404 ns 150823 ns 39560 batches_per_second=57.4566k/s rows_per_second=57.4566G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:1000/real_time/threads:1 171466 ns 168896 ns 4090 batches_per_second=5.83206M/s rows_per_second=5.83206G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:1000/real_time/threads:10 24453 ns 215813 ns 29250 batches_per_second=40.8955M/s rows_per_second=40.8955G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:10000/real_time/threads:1 174988 ns 170606 ns 4062 batches_per_second=571.467k/s rows_per_second=5.71467G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:10000/real_time/threads:10 24745 ns 219122 ns 27890 batches_per_second=4.04122M/s rows_per_second=40.4122G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:100000/real_time/threads:1 171394 ns 168804 ns 4080 batches_per_second=58.3451k/s rows_per_second=5.83451G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:100000/real_time/threads:10 23931 ns 218870 ns 28940 batches_per_second=417.863k/s rows_per_second=41.7863G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:1 170317 ns 167983 ns 4126 batches_per_second=5.8714k/s rows_per_second=5.8714G/s
SimplifyFilterWithGuarantee/positive_filter_simple_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:10 23850 ns 219822 ns 29090 batches_per_second=41.9283k/s rows_per_second=41.9283G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:1000/real_time/threads:1 216912 ns 215671 ns 3244 batches_per_second=4.61016M/s rows_per_second=4.61016G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:1000/real_time/threads:10 37816 ns 254972 ns 19270 batches_per_second=26.4435M/s rows_per_second=26.4435G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:10000/real_time/threads:1 212547 ns 209974 ns 3276 batches_per_second=470.484k/s rows_per_second=4.70484G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:10000/real_time/threads:10 29616 ns 273444 ns 23480 batches_per_second=3.37653M/s rows_per_second=33.7653G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:100000/real_time/threads:1 213545 ns 210858 ns 3280 batches_per_second=46.8285k/s rows_per_second=4.68285G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:100000/real_time/threads:10 29943 ns 272780 ns 23020 batches_per_second=333.968k/s rows_per_second=33.3968G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:1 213744 ns 211567 ns 3284 batches_per_second=4.6785k/s rows_per_second=4.6785G/s
SimplifyFilterWithGuarantee/positive_filter_cast_guarantee_dictionary/rows_per_batch:1000000/real_time/threads:10 29760 ns 270189 ns 21170 batches_per_second=33.6024k/s rows_per_second=33.6024G/s
BindAndEvaluate/simple_array/rows_per_batch:1000/real_time/threads:1 4084 ns 4044 ns 170666 batches_per_second=244.863M/s rows_per_second=244.863G/s
BindAndEvaluate/simple_array/rows_per_batch:1000/real_time/threads:10 542 ns 4899 ns 1313360 batches_per_second=1.8444G/s rows_per_second=1.8444T/s
BindAndEvaluate/simple_array/rows_per_batch:10000/real_time/threads:1 4083 ns 4029 ns 171396 batches_per_second=24.4926M/s rows_per_second=244.926G/s
BindAndEvaluate/simple_array/rows_per_batch:10000/real_time/threads:10 541 ns 4974 ns 1291170 batches_per_second=184.781M/s rows_per_second=1.84781T/s
BindAndEvaluate/simple_array/rows_per_batch:100000/real_time/threads:1 4078 ns 4039 ns 171360 batches_per_second=2.45224M/s rows_per_second=245.224G/s
BindAndEvaluate/simple_array/rows_per_batch:100000/real_time/threads:10 543 ns 4914 ns 1281100 batches_per_second=18.4217M/s rows_per_second=1.84217T/s
BindAndEvaluate/simple_array/rows_per_batch:1000000/real_time/threads:1 4081 ns 4033 ns 170748 batches_per_second=245.066k/s rows_per_second=245.066G/s
BindAndEvaluate/simple_array/rows_per_batch:1000000/real_time/threads:10 539 ns 4973 ns 1319040 batches_per_second=1.85578M/s rows_per_second=1.85578T/s
BindAndEvaluate/simple_scalar/rows_per_batch:1000/real_time/threads:1 4050 ns 4001 ns 172091 batches_per_second=246.906M/s rows_per_second=246.906G/s
BindAndEvaluate/simple_scalar/rows_per_batch:1000/real_time/threads:10 542 ns 4838 ns 1355430 batches_per_second=1.84554G/s rows_per_second=1.84554T/s
BindAndEvaluate/simple_scalar/rows_per_batch:10000/real_time/threads:1 4048 ns 4000 ns 171848 batches_per_second=24.701M/s rows_per_second=247.01G/s
BindAndEvaluate/simple_scalar/rows_per_batch:10000/real_time/threads:10 538 ns 4996 ns 1253670 batches_per_second=186.029M/s rows_per_second=1.86029T/s
BindAndEvaluate/simple_scalar/rows_per_batch:100000/real_time/threads:1 4082 ns 4021 ns 170513 batches_per_second=2.44982M/s rows_per_second=244.982G/s
BindAndEvaluate/simple_scalar/rows_per_batch:100000/real_time/threads:10 537 ns 4963 ns 1338960 batches_per_second=18.6157M/s rows_per_second=1.86157T/s
BindAndEvaluate/simple_scalar/rows_per_batch:1000000/real_time/threads:1 4086 ns 4047 ns 173320 batches_per_second=244.752k/s rows_per_second=244.752G/s
BindAndEvaluate/simple_scalar/rows_per_batch:1000000/real_time/threads:10 556 ns 4883 ns 1261530 batches_per_second=1.79886M/s rows_per_second=1.79886T/s
BindAndEvaluate/nested_array/rows_per_batch:1000/real_time/threads:1 31848 ns 31779 ns 21951 batches_per_second=31.3992M/s rows_per_second=31.3992G/s
BindAndEvaluate/nested_array/rows_per_batch:1000/real_time/threads:10 4223 ns 40172 ns 164150 batches_per_second=236.819M/s rows_per_second=236.819G/s
BindAndEvaluate/nested_array/rows_per_batch:10000/real_time/threads:1 31832 ns 31749 ns 22034 batches_per_second=3.14148M/s rows_per_second=31.4148G/s
BindAndEvaluate/nested_array/rows_per_batch:10000/real_time/threads:10 4205 ns 39491 ns 166130 batches_per_second=23.7815M/s rows_per_second=237.815G/s
BindAndEvaluate/nested_array/rows_per_batch:100000/real_time/threads:1 31901 ns 31801 ns 21994 batches_per_second=313.466k/s rows_per_second=31.3466G/s
BindAndEvaluate/nested_array/rows_per_batch:100000/real_time/threads:10 4186 ns 39733 ns 165240 batches_per_second=2.38906M/s rows_per_second=238.906G/s
BindAndEvaluate/nested_array/rows_per_batch:1000000/real_time/threads:1 31787 ns 31704 ns 21952 batches_per_second=31.459k/s rows_per_second=31.459G/s
BindAndEvaluate/nested_array/rows_per_batch:1000000/real_time/threads:10 4216 ns 40262 ns 166730 batches_per_second=237.19k/s rows_per_second=237.19G/s
BindAndEvaluate/nested_scalar/rows_per_batch:1000/real_time/threads:1 38933 ns 38814 ns 17982 batches_per_second=25.6851M/s rows_per_second=25.6851G/s
BindAndEvaluate/nested_scalar/rows_per_batch:1000/real_time/threads:10 5349 ns 50125 ns 133950 batches_per_second=186.948M/s rows_per_second=186.948G/s
BindAndEvaluate/nested_scalar/rows_per_batch:10000/real_time/threads:1 38753 ns 38669 ns 18110 batches_per_second=2.58045M/s rows_per_second=25.8045G/s
BindAndEvaluate/nested_scalar/rows_per_batch:10000/real_time/threads:10 5295 ns 49904 ns 133450 batches_per_second=18.8866M/s rows_per_second=188.866G/s
BindAndEvaluate/nested_scalar/rows_per_batch:100000/real_time/threads:1 38861 ns 38743 ns 18056 batches_per_second=257.326k/s rows_per_second=25.7326G/s
BindAndEvaluate/nested_scalar/rows_per_batch:100000/real_time/threads:10 5478 ns 50011 ns 136360 batches_per_second=1.82544M/s rows_per_second=182.544G/s
BindAndEvaluate/nested_scalar/rows_per_batch:1000000/real_time/threads:1 39104 ns 38956 ns 17992 batches_per_second=25.5726k/s rows_per_second=25.5726G/s
BindAndEvaluate/nested_scalar/rows_per_batch:1000000/real_time/threads:10 5589 ns 49032 ns 132500 batches_per_second=178.912k/s rows_per_second=178.912G/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:1000/real_time/threads:1 51566327 ns 50175286 ns 14 batches_per_second=19.3925k/s rows_per_second=19.3925M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:1000/real_time/threads:10 7677785 ns 62872600 ns 100 batches_per_second=130.246k/s rows_per_second=130.246M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:10000/real_time/threads:1 10572495 ns 10425750 ns 68 batches_per_second=9.45851k/s rows_per_second=94.5851M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:10000/real_time/threads:10 1374298 ns 12113255 ns 510 batches_per_second=72.7644k/s rows_per_second=727.644M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:100000/real_time/threads:1 6680194 ns 6537132 ns 106 batches_per_second=1.49696k/s rows_per_second=149.696M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:100000/real_time/threads:10 853062 ns 7308340 ns 900 batches_per_second=11.7225k/s rows_per_second=1.17225G/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:1000000/real_time/threads:1 6285523 ns 6261730 ns 111 batches_per_second=159.096/s rows_per_second=159.096M/s
ExecuteScalarExpressionOverhead/complex_integer_expression/rows_per_batch:1000000/real_time/threads:10 764669 ns 6859221 ns 870 batches_per_second=1.30775k/s rows_per_second=1.30775G/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:1000/real_time/threads:1 48284687 ns 48109929 ns 14 batches_per_second=20.7105k/s rows_per_second=20.7105M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:1000/real_time/threads:10 6832892 ns 62057060 ns 100 batches_per_second=146.351k/s rows_per_second=146.351M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:10000/real_time/threads:1 10248537 ns 10218868 ns 68 batches_per_second=9.75749k/s rows_per_second=97.5749M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:10000/real_time/threads:10 1346412 ns 12029128 ns 530 batches_per_second=74.2715k/s rows_per_second=742.715M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:100000/real_time/threads:1 6517764 ns 6463095 ns 105 batches_per_second=1.53427k/s rows_per_second=153.427M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:100000/real_time/threads:10 876288 ns 7263865 ns 800 batches_per_second=11.4118k/s rows_per_second=1.14118G/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:1000000/real_time/threads:1 6235951 ns 6204080 ns 112 batches_per_second=160.36/s rows_per_second=160.36M/s
ExecuteScalarExpressionOverhead/complex_expression/rows_per_batch:1000000/real_time/threads:10 767536 ns 6863618 ns 950 batches_per_second=1.30287k/s rows_per_second=1.30287G/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:1000/real_time/threads:1 5921275 ns 5854695 ns 118 batches_per_second=168.883k/s rows_per_second=168.883M/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:1000/real_time/threads:10 714197 ns 6481899 ns 1000 batches_per_second=1.40017M/s rows_per_second=1.40017G/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:10000/real_time/threads:1 6110309 ns 6079175 ns 114 batches_per_second=16.3658k/s rows_per_second=163.658M/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:10000/real_time/threads:10 819205 ns 6796318 ns 980 batches_per_second=122.07k/s rows_per_second=1.2207G/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:100000/real_time/threads:1 6158970 ns 6104316 ns 114 batches_per_second=1.62365k/s rows_per_second=162.365M/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:100000/real_time/threads:10 731520 ns 6751456 ns 970 batches_per_second=13.6702k/s rows_per_second=1.36702G/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:1000000/real_time/threads:1 6147726 ns 6115526 ns 114 batches_per_second=162.662/s rows_per_second=162.662M/s
ExecuteScalarExpressionBaseline<ComplexExpressionBaseline>/rows_per_batch:1000000/real_time/threads:10 781598 ns 6790184 ns 960 batches_per_second=1.27943k/s rows_per_second=1.27943G/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:1000/real_time/threads:1 17077564 ns 16973488 ns 41 batches_per_second=58.5564k/s rows_per_second=58.5564M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:1000/real_time/threads:10 2271629 ns 21239228 ns 290 batches_per_second=440.213k/s rows_per_second=440.213M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:10000/real_time/threads:1 8242510 ns 8176506 ns 87 batches_per_second=12.1322k/s rows_per_second=121.322M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:10000/real_time/threads:10 1330541 ns 11598270 ns 530 batches_per_second=75.1574k/s rows_per_second=751.574M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:100000/real_time/threads:1 5963924 ns 5938670 ns 115 batches_per_second=1.67675k/s rows_per_second=167.675M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:100000/real_time/threads:10 844967 ns 7430600 ns 790 batches_per_second=11.8348k/s rows_per_second=1.18348G/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:1000000/real_time/threads:1 5852841 ns 5800017 ns 121 batches_per_second=170.857/s rows_per_second=170.857M/s
ExecuteScalarExpressionOverhead/simple_expression/rows_per_batch:1000000/real_time/threads:10 845837 ns 7217863 ns 820 batches_per_second=1.18226k/s rows_per_second=1.18226G/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:1000/real_time/threads:1 2582312 ns 2574918 ns 269 batches_per_second=387.25k/s rows_per_second=387.25M/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:1000/real_time/threads:10 317662 ns 2892149 ns 2260 batches_per_second=3.148M/s rows_per_second=3.148G/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:10000/real_time/threads:1 2635967 ns 2624947 ns 265 batches_per_second=37.9367k/s rows_per_second=379.367M/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:10000/real_time/threads:10 333883 ns 2932247 ns 2180 batches_per_second=299.506k/s rows_per_second=2.99506G/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:100000/real_time/threads:1 2771566 ns 2652871 ns 263 batches_per_second=3.60807k/s rows_per_second=360.807M/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:100000/real_time/threads:10 335136 ns 2910181 ns 2200 batches_per_second=29.8386k/s rows_per_second=2.98386G/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:1000000/real_time/threads:1 2750284 ns 2722813 ns 256 batches_per_second=363.599/s rows_per_second=363.599M/s
ExecuteScalarExpressionBaseline<SimpleExpressionBaseline>/rows_per_batch:1000000/real_time/threads:10 342752 ns 2950926 ns 2080 batches_per_second=2.91756k/s rows_per_second=2.91756G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:1000/real_time/threads:1 8790873 ns 8769237 ns 80 batches_per_second=113.754k/s rows_per_second=113.754M/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:1000/real_time/threads:10 1342398 ns 12345491 ns 550 batches_per_second=744.936k/s rows_per_second=744.936M/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:10000/real_time/threads:1 872666 ns 870413 ns 796 batches_per_second=114.591k/s rows_per_second=1.14591G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:10000/real_time/threads:10 131742 ns 1269415 ns 5330 batches_per_second=759.059k/s rows_per_second=7.59059G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:100000/real_time/threads:1 87289 ns 87082 ns 8056 batches_per_second=114.562k/s rows_per_second=11.4562G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:100000/real_time/threads:10 13422 ns 125458 ns 52000 batches_per_second=745.029k/s rows_per_second=74.5029G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:1000000/real_time/threads:1 8795 ns 8769 ns 79333 batches_per_second=113.698k/s rows_per_second=113.698G/s
ExecuteScalarExpressionOverhead/zero_copy_expression/rows_per_batch:1000000/real_time/threads:10 1365 ns 12879 ns 502540 batches_per_second=732.391k/s rows_per_second=732.391G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:1000/real_time/threads:1 472178 ns 471268 ns 1474 batches_per_second=2.11784M/s rows_per_second=2.11784G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:1000/real_time/threads:10 61248 ns 568913 ns 12020 batches_per_second=16.3271M/s rows_per_second=16.3271G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:10000/real_time/threads:1 46905 ns 46806 ns 14883 batches_per_second=2.13196M/s rows_per_second=21.3196G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:10000/real_time/threads:10 6838 ns 55928 ns 121200 batches_per_second=14.6235M/s rows_per_second=146.235G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:100000/real_time/threads:1 4721 ns 4690 ns 149244 batches_per_second=2.11831M/s rows_per_second=211.831G/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:100000/real_time/threads:10 587 ns 5367 ns 1233750 batches_per_second=17.0482M/s rows_per_second=1.70482T/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:1000000/real_time/threads:1 476 ns 475 ns 1454399 batches_per_second=2.10203M/s rows_per_second=2.10203T/s
ExecuteScalarExpressionOverhead/ref_only_expression/rows_per_batch:1000000/real_time/threads:10 59.6 ns 546 ns 12177240 batches_per_second=16.7797M/s rows_per_second=16.7797T/s
```
</details>
#### filter
<details><summary>result</summary>
```
Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
***WARNING*** Failed to set thread affinity. Estimated CPU frequency may be incorrect.
2024-03-29T19:02:27+09:00
Running /Users/lama/workspace/arrow-new/cpp/cmake-build-debug/debug/arrow-acero-filter-benchmark
Run on (10 X 24 MHz CPU s)
CPU Caches:
L1 Data 64 KiB
L1 Instruction 128 KiB
L2 Unified 4096 KiB (x10)
Load Average: 21.65, 21.11, 17.51
***WARNING*** Library was built as DEBUG. Timings may be affected.
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations UserCounters...
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
FilterOverheadIsolated/complex_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 32219871 ns 7680273 ns 22 batches_per_second=31.0367k/s rows_per_second=31.0367M/s
FilterOverheadIsolated/complex_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 9548148 ns 1234627 ns 75 batches_per_second=10.4732k/s rows_per_second=104.732M/s
FilterOverheadIsolated/complex_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 9220167 ns 358654 ns 81 batches_per_second=1.08458k/s rows_per_second=108.458M/s
FilterOverheadIsolated/complex_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 47116182 ns 360786 ns 14 batches_per_second=21.2241/s rows_per_second=21.2241M/s
FilterOverheadIsolated/simple_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 26449385 ns 7705923 ns 26 batches_per_second=37.8081k/s rows_per_second=37.8081M/s
FilterOverheadIsolated/simple_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 7721969 ns 1225149 ns 94 batches_per_second=12.9501k/s rows_per_second=129.501M/s
FilterOverheadIsolated/simple_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 7580444 ns 346980 ns 98 batches_per_second=1.31918k/s rows_per_second=131.918M/s
FilterOverheadIsolated/simple_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 36332956 ns 227526 ns 19 batches_per_second=27.5232/s rows_per_second=27.5232M/s
FilterOverheadIsolated/ref_only_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 19977948 ns 7883914 ns 35 batches_per_second=50.0552k/s rows_per_second=50.0552M/s
FilterOverheadIsolated/ref_only_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 5588740 ns 1262710 ns 124 batches_per_second=17.8931k/s rows_per_second=178.931M/s
FilterOverheadIsolated/ref_only_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 5455305 ns 364585 ns 130 batches_per_second=1.83308k/s rows_per_second=183.308M/s
FilterOverheadIsolated/ref_only_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 27660087 ns 225600 ns 25 batches_per_second=36.1532/s rows_per_second=36.1532M/s
FilterOverhead/complex_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 37483877 ns 5350263 ns 19 batches_per_second=26.6781k/s rows_per_second=26.6781M/s
FilterOverhead/complex_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 11208230 ns 956475 ns 61 batches_per_second=8.92202k/s rows_per_second=89.2202M/s
FilterOverhead/complex_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 12460321 ns 611911 ns 56 batches_per_second=802.548/s rows_per_second=80.2548M/s
FilterOverhead/complex_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 65357705 ns 487455 ns 11 batches_per_second=15.3004/s rows_per_second=15.3004M/s
FilterOverhead/simple_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 33649222 ns 5427714 ns 21 batches_per_second=29.7184k/s rows_per_second=29.7184M/s
FilterOverhead/simple_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 9939710 ns 974211 ns 71 batches_per_second=10.0607k/s rows_per_second=100.607M/s
FilterOverhead/simple_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 10684081 ns 606750 ns 64 batches_per_second=935.972/s rows_per_second=93.5972M/s
FilterOverhead/simple_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 55282253 ns 448538 ns 13 batches_per_second=18.089/s rows_per_second=18.089M/s
FilterOverhead/ref_only_expression/batch_size:1000/null_prob:0/bool_true_prob:50/real_time 28750612 ns 5398200 ns 25 batches_per_second=34.7819k/s rows_per_second=34.7819M/s
FilterOverhead/ref_only_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time 8271133 ns 981464 ns 84 batches_per_second=12.0902k/s rows_per_second=120.902M/s
FilterOverhead/ref_only_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time 8353045 ns 591232 ns 82 batches_per_second=1.19717k/s rows_per_second=119.717M/s
FilterOverhead/ref_only_expression/batch_size:1000000/null_prob:0/bool_true_prob:50/real_time 43055520 ns 572294 ns 17 batches_per_second=23.2258/s rows_per_second=23.2258M/s
FilterOverhead/selectivity_benchmark/batch_size:1000/null_prob:10/bool_true_prob:50/real_time 34575329 ns 8412050 ns 20 batches_per_second=28.9224k/s rows_per_second=28.9224M/s
FilterOverhead/selectivity_benchmark/batch_size:1000/null_prob:50/bool_true_prob:50/real_time 33501100 ns 7702550 ns 20 batches_per_second=29.8498k/s rows_per_second=29.8498M/s
FilterOverhead/selectivity_benchmark/batch_size:1000/null_prob:75/bool_true_prob:50/real_time 30975611 ns 7026435 ns 23 batches_per_second=32.2835k/s rows_per_second=32.2835M/s
FilterOverhead/selectivity_benchmark/batch_size:1000/null_prob:100/bool_true_prob:50/real_time 24824322 ns 1811310 ns 29 batches_per_second=40.2831k/s rows_per_second=40.2831M/s
FilterOverhead/selectivity_benchmark/batch_size:10000/null_prob:10/bool_true_prob:50/real_time 7229531 ns 1575812 ns 96 batches_per_second=13.8322k/s rows_per_second=138.322M/s
FilterOverhead/selectivity_benchmark/batch_size:10000/null_prob:50/bool_true_prob:50/real_time 6805925 ns 1223817 ns 104 batches_per_second=14.6931k/s rows_per_second=146.931M/s
FilterOverhead/selectivity_benchmark/batch_size:10000/null_prob:75/bool_true_prob:50/real_time 5268858 ns 932575 ns 127 batches_per_second=18.9794k/s rows_per_second=189.794M/s
FilterOverhead/selectivity_benchmark/batch_size:10000/null_prob:100/bool_true_prob:50/real_time 2763840 ns 254522 ns 251 batches_per_second=36.1815k/s rows_per_second=361.815M/s
FilterOverhead/selectivity_benchmark/batch_size:100000/null_prob:10/bool_true_prob:50/real_time 6488062 ns 853818 ns 88 batches_per_second=1.54129k/s rows_per_second=154.129M/s
FilterOverhead/selectivity_benchmark/batch_size:100000/null_prob:50/bool_true_prob:50/real_time 6425266 ns 715573 ns 110 batches_per_second=1.55636k/s rows_per_second=155.636M/s
FilterOverhead/selectivity_benchmark/batch_size:100000/null_prob:75/bool_true_prob:50/real_time 4615714 ns 616526 ns 152 batches_per_second=2.16651k/s rows_per_second=216.651M/s
FilterOverhead/selectivity_benchmark/batch_size:100000/null_prob:100/bool_true_prob:50/real_time 1443950 ns 129039 ns 463 batches_per_second=6.92545k/s rows_per_second=692.545M/s
FilterOverhead/selectivity_benchmark/batch_size:1000000/null_prob:10/bool_true_prob:50/real_time 30118737 ns 830696 ns 23 batches_per_second=33.2019/s rows_per_second=33.2019M/s
FilterOverhead/selectivity_benchmark/batch_size:1000000/null_prob:50/bool_true_prob:50/real_time 30282326 ns 594652 ns 23 batches_per_second=33.0226/s rows_per_second=33.0226M/s
FilterOverhead/selectivity_benchmark/batch_size:1000000/null_prob:75/bool_true_prob:50/real_time 19982260 ns 478914 ns 35 batches_per_second=50.0444/s rows_per_second=50.0444M/s
FilterOverhead/selectivity_benchmark/batch_size:1000000/null_prob:100/bool_true_prob:50/real_time 4219737 ns 93421 ns 164 batches_per_second=236.982/s rows_per_second=236.982M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000/null_prob:0/bool_true_prob:25/real_time 41057035 ns 4822889 ns 18 batches_per_second=24.3564k/s rows_per_second=24.3564M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000/null_prob:25/bool_true_prob:33/real_time 45722763 ns 7106687 ns 16 batches_per_second=21.8709k/s rows_per_second=21.8709M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000/null_prob:50/bool_true_prob:50/real_time 43468094 ns 7015125 ns 16 batches_per_second=23.0054k/s rows_per_second=23.0054M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000/null_prob:75/bool_true_prob:100/real_time 41803265 ns 6983588 ns 17 batches_per_second=23.9216k/s rows_per_second=23.9216M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:10000/null_prob:0/bool_true_prob:25/real_time 8723163 ns 709687 ns 83 batches_per_second=11.4637k/s rows_per_second=114.637M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:10000/null_prob:25/bool_true_prob:33/real_time 10929365 ns 914864 ns 66 batches_per_second=9.14966k/s rows_per_second=91.4966M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:10000/null_prob:50/bool_true_prob:50/real_time 10055967 ns 956829 ns 70 batches_per_second=9.94434k/s rows_per_second=99.4434M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:10000/null_prob:75/bool_true_prob:100/real_time 6704041 ns 915833 ns 102 batches_per_second=14.9164k/s rows_per_second=149.164M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:100000/null_prob:0/bool_true_prob:25/real_time 8595726 ns 534810 ns 84 batches_per_second=1.16337k/s rows_per_second=116.337M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:100000/null_prob:25/bool_true_prob:33/real_time 11354318 ns 650177 ns 62 batches_per_second=880.722/s rows_per_second=88.0722M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:100000/null_prob:50/bool_true_prob:50/real_time 10747181 ns 655785 ns 65 batches_per_second=930.477/s rows_per_second=93.0477M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:100000/null_prob:75/bool_true_prob:100/real_time 5781438 ns 623121 ns 116 batches_per_second=1.72967k/s rows_per_second=172.967M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000000/null_prob:0/bool_true_prob:25/real_time 42612688 ns 448313 ns 16 batches_per_second=23.4672/s rows_per_second=23.4672M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000000/null_prob:25/bool_true_prob:33/real_time 54707724 ns 522846 ns 13 batches_per_second=18.279/s rows_per_second=18.279M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000000/null_prob:50/bool_true_prob:50/real_time 50207524 ns 486357 ns 14 batches_per_second=19.9173/s rows_per_second=19.9173M/s
FilterOverhead/not_null_to_is_true_multipass_benchmark/batch_size:1000000/null_prob:75/bool_true_prob:100/real_time 26317915 ns 551704 ns 27 batches_per_second=37.9969/s rows_per_second=37.9969M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000/null_prob:0/bool_true_prob:25/real_time 38054033 ns 7078632 ns 19 batches_per_second=26.2784k/s rows_per_second=26.2784M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000/null_prob:25/bool_true_prob:33/real_time 38641044 ns 7008333 ns 18 batches_per_second=25.8792k/s rows_per_second=25.8792M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000/null_prob:50/bool_true_prob:50/real_time 39134701 ns 6840667 ns 18 batches_per_second=25.5528k/s rows_per_second=25.5528M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000/null_prob:75/bool_true_prob:100/real_time 39117280 ns 7301111 ns 18 batches_per_second=25.5641k/s rows_per_second=25.5641M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:10000/null_prob:0/bool_true_prob:25/real_time 7060370 ns 962270 ns 100 batches_per_second=14.1636k/s rows_per_second=141.636M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:10000/null_prob:25/bool_true_prob:33/real_time 7138369 ns 947133 ns 98 batches_per_second=14.0088k/s rows_per_second=140.088M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:10000/null_prob:50/bool_true_prob:50/real_time 7310957 ns 947340 ns 97 batches_per_second=13.6781k/s rows_per_second=136.781M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:10000/null_prob:75/bool_true_prob:100/real_time 7301242 ns 949224 ns 98 batches_per_second=13.6963k/s rows_per_second=136.963M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:100000/null_prob:0/bool_true_prob:25/real_time 6378482 ns 619629 ns 105 batches_per_second=1.56777k/s rows_per_second=156.777M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:100000/null_prob:25/bool_true_prob:33/real_time 6775960 ns 623058 ns 104 batches_per_second=1.47581k/s rows_per_second=147.581M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:100000/null_prob:50/bool_true_prob:50/real_time 6766374 ns 618680 ns 100 batches_per_second=1.4779k/s rows_per_second=147.79M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:100000/null_prob:75/bool_true_prob:100/real_time 6795010 ns 634635 ns 104 batches_per_second=1.47167k/s rows_per_second=147.167M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000000/null_prob:0/bool_true_prob:25/real_time 30435152 ns 537565 ns 23 batches_per_second=32.8567/s rows_per_second=32.8567M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000000/null_prob:25/bool_true_prob:33/real_time 31719311 ns 488864 ns 22 batches_per_second=31.5265/s rows_per_second=31.5265M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000000/null_prob:50/bool_true_prob:50/real_time 31834165 ns 503318 ns 22 batches_per_second=31.4128/s rows_per_second=31.4128M/s
FilterOverhead/not_null_and_is_true_singlepass_benchmark/batch_size:1000000/null_prob:75/bool_true_prob:100/real_time 31745786 ns 489682 ns 22 batches_per_second=31.5002/s rows_per_second=31.5002M/s
```
</details>
#### project
<details><summary>result</summary>
```
Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
***WARNING*** Failed to set thread affinity. Estimated CPU frequency may be incorrect.
2024-03-29T19:04:07+09:00
Running /Users/lama/workspace/arrow-new/cpp/cmake-build-debug/debug/arrow-acero-project-benchmark
Run on (10 X 24 MHz CPU s)
CPU Caches:
L1 Data 64 KiB
L1 Instruction 128 KiB
L2 Unified 4096 KiB (x10)
Load Average: 9.33, 16.86, 16.25
***WARNING*** Library was built as DEBUG. Timings may be affected.
---------------------------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations UserCounters...
---------------------------------------------------------------------------------------------------------------------------------------
ProjectionOverheadIsolated/complex_expression/batch_size:1000/real_time 16315196 ns 3740581 ns 43 batches_per_second=61.2926k/s rows_per_second=61.2926M/s
ProjectionOverheadIsolated/complex_expression/batch_size:10000/real_time 2475621 ns 469714 ns 280 batches_per_second=40.3939k/s rows_per_second=403.939M/s
ProjectionOverheadIsolated/complex_expression/batch_size:100000/real_time 1549914 ns 123036 ns 440 batches_per_second=6.45197k/s rows_per_second=645.197M/s
ProjectionOverheadIsolated/complex_expression/batch_size:1000000/real_time 6909512 ns 51307 ns 101 batches_per_second=144.728/s rows_per_second=144.728M/s
ProjectionOverheadIsolated/simple_expression/batch_size:1000/real_time 11964503 ns 6325190 ns 58 batches_per_second=83.5806k/s rows_per_second=83.5806M/s
ProjectionOverheadIsolated/simple_expression/batch_size:10000/real_time 2716515 ns 1321031 ns 257 batches_per_second=36.8119k/s rows_per_second=368.119M/s
ProjectionOverheadIsolated/simple_expression/batch_size:100000/real_time 1773873 ns 489173 ns 393 batches_per_second=5.63738k/s rows_per_second=563.738M/s
ProjectionOverheadIsolated/simple_expression/batch_size:1000000/real_time 5827351 ns 251849 ns 119 batches_per_second=171.605/s rows_per_second=171.605M/s
ProjectionOverheadIsolated/zero_copy_expression/batch_size:1000/real_time 5752426 ns 1268032 ns 124 batches_per_second=173.84k/s rows_per_second=173.84M/s
ProjectionOverheadIsolated/zero_copy_expression/batch_size:10000/real_time 638638 ns 186231 ns 1067 batches_per_second=156.583k/s rows_per_second=1.56583G/s
ProjectionOverheadIsolated/zero_copy_expression/batch_size:100000/real_time 125635 ns 61002 ns 5414 batches_per_second=79.5956k/s rows_per_second=7.95956G/s
ProjectionOverheadIsolated/zero_copy_expression/batch_size:1000000/real_time 38479 ns 14166 ns 18069 batches_per_second=25.988k/s rows_per_second=25.988G/s
ProjectionOverheadIsolated/ref_only_expression/batch_size:1000/real_time 4093881 ns 884006 ns 171 batches_per_second=244.267k/s rows_per_second=244.267M/s
ProjectionOverheadIsolated/ref_only_expression/batch_size:10000/real_time 488240 ns 144868 ns 1444 batches_per_second=204.817k/s rows_per_second=2.04817G/s
ProjectionOverheadIsolated/ref_only_expression/batch_size:100000/real_time 80669 ns 37637 ns 7370 batches_per_second=123.964k/s rows_per_second=12.3964G/s
ProjectionOverheadIsolated/ref_only_expression/batch_size:1000000/real_time 24901 ns 13072 ns 27699 batches_per_second=40.1594k/s rows_per_second=40.1594G/s
ProjectionOverhead/complex_expression/batch_size:1000/real_time 24644825 ns 3193034 ns 29 batches_per_second=40.5765k/s rows_per_second=40.5765M/s
ProjectionOverhead/complex_expression/batch_size:10000/real_time 3520050 ns 447907 ns 193 batches_per_second=28.4087k/s rows_per_second=284.087M/s
ProjectionOverhead/complex_expression/batch_size:100000/real_time 2530539 ns 287728 ns 279 batches_per_second=3.95173k/s rows_per_second=395.173M/s
ProjectionOverhead/complex_expression/batch_size:1000000/real_time 9870693 ns 250451 ns 71 batches_per_second=101.31/s rows_per_second=101.31M/s
ProjectionOverhead/simple_expression/batch_size:1000/real_time 24051871 ns 5912379 ns 29 batches_per_second=41.5768k/s rows_per_second=41.5768M/s
ProjectionOverhead/simple_expression/batch_size:10000/real_time 3486256 ns 1254348 ns 201 batches_per_second=28.6841k/s rows_per_second=286.841M/s
ProjectionOverhead/simple_expression/batch_size:100000/real_time 2386357 ns 696672 ns 299 batches_per_second=4.19049k/s rows_per_second=419.049M/s
ProjectionOverhead/simple_expression/batch_size:1000000/real_time 7903190 ns 664575 ns 87 batches_per_second=126.531/s rows_per_second=126.531M/s
ProjectionOverhead/zero_copy_expression/batch_size:1000/real_time 18249752 ns 743842 ns 38 batches_per_second=54.7953k/s rows_per_second=54.7953M/s
ProjectionOverhead/zero_copy_expression/batch_size:10000/real_time 1941304 ns 132713 ns 363 batches_per_second=51.5118k/s rows_per_second=515.118M/s
ProjectionOverhead/zero_copy_expression/batch_size:100000/real_time 765708 ns 84303 ns 908 batches_per_second=13.0598k/s rows_per_second=1.30598G/s
ProjectionOverhead/zero_copy_expression/batch_size:1000000/real_time 1193561 ns 75816 ns 625 batches_per_second=837.829/s rows_per_second=837.829M/s
ProjectionOverhead/ref_only_expression/batch_size:1000/real_time 18194863 ns 716585 ns 41 batches_per_second=54.9606k/s rows_per_second=54.9606M/s
ProjectionOverhead/ref_only_expression/batch_size:10000/real_time 1893367 ns 122503 ns 374 batches_per_second=52.816k/s rows_per_second=528.16M/s
ProjectionOverhead/ref_only_expression/batch_size:100000/real_time 691819 ns 73757 ns 1002 batches_per_second=14.4547k/s rows_per_second=1.44547G/s
ProjectionOverhead/ref_only_expression/batch_size:1000000/real_time 771102 ns 61396 ns 934 batches_per_second=1.29685k/s rows_per_second=1.29685G/s
```
</details>
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "westonpace (via GitHub)" <gi...@apache.org>.
westonpace merged PR #40884:
URL: https://github.com/apache/arrow/pull/40884
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "llama90 (via GitHub)" <gi...@apache.org>.
llama90 commented on code in PR #40884:
URL: https://github.com/apache/arrow/pull/40884#discussion_r1573119356
##########
cpp/src/arrow/acero/asof_join_benchmark.cc:
##########
@@ -91,7 +91,7 @@ static void TableJoinOverhead(benchmark::State& state,
ASSERT_OK(DeclarationToStatus(std::move(join_node), /*use_threads=*/false));
}
- state.counters["input_rows_per_second"] = benchmark::Counter(
+ state.counters["rows_per_second"] = benchmark::Counter(
Review Comment:
All the other codes also use the `rows_per_second` value, so I made the modifications accordingly.
--
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] GH-39664: [C++][Acero] Ensure Acero benchmarks present a metric for identifying throughput [arrow]
Posted by "conbench-apache-arrow[bot] (via GitHub)" <gi...@apache.org>.
conbench-apache-arrow[bot] commented on PR #40884:
URL: https://github.com/apache/arrow/pull/40884#issuecomment-2070614969
After merging your PR, Conbench analyzed the 7 benchmarking runs that have been run so far on merge-commit 79799e59b1a0c7bde8b85ea955593b2d63d2a46c.
There were no benchmark performance regressions. 🎉
The [full Conbench report](https://github.com/apache/arrow/runs/24118590595) has more details. It also includes information about 7 possible false positives for unstable benchmarks that are known to sometimes produce them.
--
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