WillAyd commented on code in PR #14938:
URL: https://github.com/apache/arrow/pull/14938#discussion_r1047667864
##########
cpp/src/arrow/compute/kernels/vector_sort_benchmark.cc:
##########
@@ -301,5 +333,38 @@ BENCHMARK(TableSortIndicesInt64Wide)
})
->Unit(benchmark::TimeUnit::kNanosecond);
+//
+// Rank benchmark declarations
+//
+
+void ArrayRankSetArgs(benchmark::internal::Benchmark* bench) {
+ // 3 benchmark arguments: size, inverse null proportion, rank tiebreaker
+ bench->Unit(benchmark::kNanosecond);
+ bench->ArgNames({"", "", "tiebreaker"});
+
+ // Use only a subset of kInverseNullProportions as the cartesian product of
+ // arguments is large already.
+ const std::vector<ArgsType> inverse_null_proportions{10, 1, 0};
+ // Don't bother with Max as it should have the same perf as Min
Review Comment:
Not sure if worth commenting but I'm guessing the code implicitly assumes
the same about first / last performance
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
