yordan-pavlov commented on issue #200: URL: https://github.com/apache/arrow-rs/issues/200#issuecomment-860275967
I finally had some time to check how the new `ArrowArrayReader` affects TPC-H benchmark results - for queries which use string columns (queries 1 and 12), there is a performance improvement of about 30%, other queries that I tested, which mostly use non-string columns are unaffected. This makes sense as the new `ArrowArrayReader` is only enabled for string arrays currently. Here are the results: **before new ArrowArrayReader:** Query 1 avg time: 822.14 ms Query 3 avg time: 432.85 ms Query 5 avg time: 698.90 ms Query 6 avg time: 319.38 ms Query 12 avg time: 682.50 ms **after new ArrowArrayReader:** Query 1 avg time: 514.88 ms Query 3 avg time: 441.08 ms Query 5 avg time: 702.91 ms Query 6 avg time: 324.05 ms Query 12 avg time: 425.38 ms -- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
