rmdmattingly commented on PR #6543: URL: https://github.com/apache/hbase/pull/6543#issuecomment-2571353985
Build looks good here, and all of the balancer tests reliably pass on my machine within about 30min. I've improved the runtime of several tests too, because we made a lot of quick assumptions about runtime being 30s here, 60s there, and those really add up when running the full test suite repeatedly. I've also setup a large cluster test for conditional replica balancing, at an identical scale to the existing large cluster test for legacy replica balancing. It demonstrates a _significant_ improvement in balancer latency when dealing with 1k servers, 20k regions, 3 replicas per region, and 100 tables: <img width="515" alt="Screenshot 2025-01-04 at 11 48 55 AM" src="https://github.com/user-attachments/assets/15049da7-1e24-46af-971d-c22d2e07b8c5"; /> Because this PR is huge, and there is a lot of iteration along the way, I'm tempted to close this and reopen a clean PR with a passing build. I'll do that either this weekend, or early next week. -- 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: issues-unsubscr...@hbase.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
