wirybeaver opened a new pull request, #12977: URL: https://github.com/apache/pinot/pull/12977
Intuition: Reduce repeated byte array allocation on jvm heap when QPS is high. Is it thread safe? Yes, since the ReaderContext is created per SegmentPlanNode per query. A SegmentPlanNode would only be executed by a query worker. Additional Memory cost: number of query workers * _maxReusableByteSize (should be set as a fieldConfig per table). In the future, it will be number of query workers * min(_longestLengthOfEntry, _maxReusableByteSize) if we can add _longestLengthOfEntry in the Table V5 header. The proof demonstrating the benefits of reusable byte array on V2 / V3: -- 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]
