appletreeisyellow commented on issue #11170: URL: https://github.com/apache/datafusion/issues/11170#issuecomment-2276692323
I was trying to see where the time was spent for `FixedSizeBinary`, so I used [Mac's Instrument CPU Profiler](https://developer.apple.com/videos/play/wwdc2019/411/) to run a CPU profiling on the benchmark / example code: [samuelcolvin/datafusion-id-experiment](https://github.com/samuelcolvin/datafusion-id-experiment) This is what I observed: 98% of the time was spent on reading data from parquet files and 1.2% of the time was spent on querying (see screenshot below). So it is hard to tell whether the unsorted `FixSizeBinary` is slower or not with the current example code. Maybe other people can try a different approach to measure where the time was spent -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
