Through this blog I see that starting from 8.3, we have significantly reduced Usage of each shard of the heap. I would like to know more details about how we can reduce the heap usage per shard. Anyone knows a PR or blog about this optimization, i can't find it in release note of 8.3
It was a combination of lots of small fixes, several of which landed in versions before 8.3 too. See Fix Large Shard Count Scalability Issues · Issue #77466 · elastic/elasticsearch · GitHub.
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.