alamb commented on PR #9277: URL: https://github.com/apache/arrow-rs/pull/9277#issuecomment-3810917158
> I don't know how much I can trust the results as it says: > > ``` > take bool null values null indices 1024 1.00 2.1±0.03µs ? ?/sec 1.78 3.6±0.08µs ? ?/sec > ``` > > almost 2 times faster for a branch of code that I did not touch. > > (it can be possible that this improved due to instruction locality but I don't know yeah I agree all of these numbers need to be carefully reviewed. I also found that for benchmarks such as this which execute in µs, the allocation pattern / allocator can cause substantial changes in the performance if other benchmarks do a different allocation pattern, for example -- 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]
