iamsanjay commented on PR #2612: URL: https://github.com/apache/solr/pull/2612#issuecomment-2376310523
Latest Benchmark with different values for maxConnectionPerDestination ``` Benchmark (maxConnectionPerHost) (nodeCount) (numReplicas) (numShards) (preGenerate) (useSmallDocs) Mode Cnt Score Error Units CloudIndexing.indexDoc 1 4 3 4 50000 false thrpt 4 1118.424 ± 847.096 ops/s CloudIndexing.indexDoc 2 4 3 4 50000 false thrpt 4 1701.146 ± 1943.553 ops/s CloudIndexing.indexDoc 4 4 3 4 50000 false thrpt 4 1726.712 ± 2448.684 ops/s CloudIndexing.indexDoc 8 4 3 4 50000 false thrpt 4 1650.018 ± 2238.615 ops/s CloudIndexing.indexDoc 16 4 3 4 50000 false thrpt 4 1651.497 ± 2117.253 ops/s CloudIndexing.indexDoc 32 4 3 4 50000 false thrpt 4 1655.557 ± 2347.372 ops/s CloudIndexing.indexDoc 64 4 3 4 50000 false thrpt 4 1695.647 ± 2139.217 ops/s CloudIndexing.indexDoc 128 4 3 4 50000 false thrpt 4 1737.919 ± 2381.445 ops/s ``` It's a good improvement as we increase maxConnectionPerStream. The pattern I see so far that If we increase the maxConnectionPerDestination then yes there are failures, but all of them are related to Out Of Memory exception rather than `cancel` stream error. Clearly, increasing `maxConnectionPerStream` also requires more memory., which is understandable. -- 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...@solr.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org For additional commands, e-mail: issues-h...@solr.apache.org