Re: [PR] PARQUET-2423: [C++][Parquet] Avoid allocating buffer object in RecordReader's SkipRecords [arrow]

["jp0317 (via GitHub)" <gi...@apache.org>]

jp0317 commented on code in PR #39818:
URL: https://github.com/apache/arrow/pull/39818#discussion_r1507784921


##########
cpp/src/parquet/column_reader_benchmark.cc:
##########
@@ -219,6 +251,20 @@ BENCHMARK(RecordReaderReadRecords)
     ->Args({2, 1000, true})
     ->Args({2, 1000, false});
 
+BENCHMARK(RecordReaderReadAndSkipRecords)
+    ->ArgNames({"Repetition", "BatchSize", "LevelsPerPage"})
+    ->Args({2, 1000, 80000})
+    ->Args({0, 1000, 80000})
+    ->Args({1, 1, 80000})

Review Comment:
   just wanted to simulate some extreme cases. I guess as long as the batch size is less than 1k it would show a noticeable perf difference



-- 
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