rondagostino commented on PR #13280: URL: https://github.com/apache/kafka/pull/13280#issuecomment-1512118780
Here are some interesting GC results, via `./jmh.sh --prof gc TopicsImageSingleRecord` showing linear GC per operation on the order of megabytes now being constant at under 2k. **Old Code** ``` Benchmark (numReplicasPerBroker) (partitionsPerTopic) (replicationFactor) (totalTopicCount) Mode Cnt Score Error Units testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 12500 avgt 5 1193832.055 ± 0.015 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 25000 avgt 5 2387008.169 ± 0.027 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 50000 avgt 5 4773480.672 ± 0.570 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 100000 avgt 5 9546305.345 ± 0.077 B/op ``` **New Code** ``` Benchmark (numReplicasPerBroker) (partitionsPerTopic) (replicationFactor) (totalTopicCount) Mode Cnt Score Error Units testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 12500 avgt 5 1744.000 ± 0.001 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 25000 avgt 5 1864.000 ± 0.001 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 50000 avgt 5 1984.000 ± 0.001 B/op testTopicsDeltaSingleTopicAdd:·gc.alloc.rate.norm 10000 10 3 100000 avgt 5 1984.000 ± 0.001 B/op ``` -- 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: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org