quantranhong1999 commented on PR #2814: URL: https://github.com/apache/james-project/pull/2814#issuecomment-3305325461
Hi @jeantil, I just ran `mvn -B -e -fae clean test -Dorg.slf4j.simpleLogger.showDateTime=true -Ddevelocity.cache.local.enabled=false` on the `james-server-webadmin-mailbox` module. (without `-Ddevelocity.cache.local.enabled=false`, the cache does indeed work for local build) Master behavior as of today (forkCount=1) => 05:56 min https://github.com/apache/james-project/pull/2814/commits/40db4b1c829893928000b786a6f34b6d6c51fb35 (forkCount=2) => 04:05 min So forkCount does indeed help more or less. > This is especially true for the forkCount changes : the result they introduce will be highly dependant on the hardware they run on, in some situations they may even have adverse effect. Merging more parallelization at the same time as merging changes that are guaranteed to have a positive impact may end up hiding a negative impact. I agree. Tuning `forkCount` to a too high value could have a negative impact on slow hardware. That is why I only increase it from 1 -> 2. That is defensive enough but still gain value IMO. > if I were doing this at work, I would keep the fork count changes in a branch for some time ( a couple of weeks ?) and setup an automatic rebase for every merge on master to accumulate datapoints on both setups and make sure the increased parallelization actually has a positive effect. In James project, we tune `forkCount` a lot: <img width="1321" height="652" alt="image" src="https://github.com/user-attachments/assets/7307622c-3f8d-4aa2-9399-1da45a37ead2"; /> So adding a few more `forkCount=2` here is safe IMO. I propose to revert it (easily) if any issue. -- 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]
