alamb commented on pull request #8635: URL: https://github.com/apache/arrow/pull/8635#issuecomment-725438655
> Can you use https://github.com/vertexclique/zor/blob/master/zor if you are on Linux? I am running on mac osx -- I looked at zor and it looked like a good tool to try and make reproducible results by clearing kernel caches, etc. I wonder if you can try using the seeded random approach with zor to see if that is as reproducible. > Yeah, I didn't want to write a const compiled seed and people forget to use the same seed every time. That is a good point -- maybe we can make a utility function or something for the tests to use. I don't think the actual value of the seed (e.g. `42`) is is important to keep consistent in all the .rs sources - just that it is a constant number between invocations of `cargo bench ...` across runs > My implementation does some extra faster stuff compared to rand too. Here: https://docs.rs/bastion-utils/0.3.2/src/bastion_utils/math.rs.html#4-28 That looks cool, but I am not enough of a numerical algorithms expert to evaluate the random ness properties of that algorithm ---------------------------------------------------------------- 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
