alamb commented on issue #7983: URL: https://github.com/apache/arrow-rs/issues/7983#issuecomment-3156054041
> [@alamb](https://github.com/alamb) Just curious, previously I worked on a system like presto/impala, which has metrics for parquet decompress time/decode time/io time. Currently in arrow-rs, seems we don't have these wrappers, how can we get the time-spent here? TLDR is there are no metrics reported by the arrow-rs parquet reader directly. I also very much would like more fine grained metrics, I actually have a PR up to propose adding such an API here - https://github.com/apache/arrow-rs/pull/8003 Perhaps you can take a look at it and let me know if that is what you had in mind > I know flamegraph would help but is there some offical way to acquire time spending on these? And how do datafusion get these metrics? DataFusion gets its metrics by observing the reader from the outside, so it can't do things like separate the decompress from decode times. -- 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-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
