fjetter commented on issue #38389: URL: https://github.com/apache/arrow/issues/38389#issuecomment-1779001544
Just a heads up. My current working theory is that the parquet deserialization performance is roughly where it is supposed to be (but honestly I don't know) but what we're seeing is rather an artifact of storage latency due to too many small reads. I'll keep that conversation separate from _this_ issue here, see https://github.com/coiled/benchmarks/issues/1125#issuecomment-1778995537 --- Still, I think the question _this_ issue is asking is whether the performance for the _deserialization_ (w/out IO) we're seeing is what is expected and if the severe slowdown on other hardware is expected. -- 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]
