countmdm opened a new pull request, #15637: URL: https://github.com/apache/lucene/pull/15637
### Description In my previous PR https://github.com/apache/lucene/pull/15618 I changed SegementTermsEnumFrame to lazily initialize byte[] arrays that it manages. After this change, we measured memory allocation of one of apps at LinkedIn that use Lucene and found that (a) its memory allocation rate went down by ~25%, as expected, but (b) still ~10% of that reduced memory allocation was due to ByteArrayDataInput objects eagerly allocated by SegmentTermsEnumFrame code. Previously we determined that a high percentage of SegmentTermsEnumFrame objects are allocated but not used by our app. Thus, the same percentage of ByteArrayDataInputs managed by STEF are wasted. This PR addresses that problem by switching to lazy allocation of these objects. -- 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]
