dnhatn opened a new pull request, #15723: URL: https://github.com/apache/lucene/pull/15723
In Elasticsearch, we execute intra-segment searches across multiple threads, which can trigger redundant, concurrent cache populations for the same segment. Since each slice-search targets a subset of the segment but cache population scans the entire segment, this amplification significantly wastes I/O and CPU for expensive queries. This PR introduces a hook in LRUQueryCache to allow custom population strategies, such as delegating to a background thread or skipping redundant populations. There is no change to the default behavior of LRUQueryCache. -- 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]
