jbewing commented on PR #5576: URL: https://github.com/apache/hbase/pull/5576#issuecomment-1865448798
Based on the most recent benchmarking results ([see JIRA comment](https://issues.apache.org/jira/browse/HBASE-28256?focusedCommentId=17799200&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-17799200)), I went and converted the approach here to use an approach adapted from [HBASE-14186](https://issues.apache.org/jira/browse/HBASE-14186). This approach has microbenchmarks that outperform our current setup. Additionally, there's no regression in the smaller vLong range like there is with the first approach that I used in this patch and performance is similar to or slightly worse in the higher vLong range than the approach I originally presented. However, after I implement [HBASE-27730](https://issues.apache.org/jira/browse/HBASE-27730), this newer approach drastically outperforms my original approach. Additionally, since there is no performance regression I think that it's safe to just replace the body of the existing readVLong method with this new implementation. Thus, this patch can be back-ported to branch-3 & branch-2. -- 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: issues-unsubscr...@hbase.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org