xiangfu0 commented on PR #19397: URL: https://github.com/apache/pinot/pull/19397#issuecomment-5488052626
Rebased and updated at `11b7ce6cfc`; all review threads are addressed and resolved. Key follow-ups: - T64 now uses sequential word-at-a-time decode and stored-type-aware retained-buffer bounds. - Plan-cache creation/eviction now has bounded concurrent canonical/alias coverage. - Cached scratch views preserve caller state, moving windows, byte order, and separate-scratch concurrency. - DELTA/DELTADELTA indexed access removes the warmed regression: matched three-fork JMH is 28.1% faster transform-only and 3.8% faster with LZ4, with allocations reduced to profiler noise. Validation: 312 codec/parser tests pass; Spotless, Checkstyle, license formatting, and license checks pass. Please re-review when convenient. -- 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]
