franz1981 commented on pull request #3566: URL: https://github.com/apache/activemq-artemis/pull/3566#issuecomment-833834162
Preliminary tests using paging won't show such regression anyway: no batching ``` ************** RUN 1 EndToEnd Throughput: 10626 ops/sec ************** EndToEnd SERVICE-TIME Latencies distribution in MICROSECONDS mean 1522.66 min 129.54 50.00% 1376.26 90.00% 1507.33 99.00% 5734.40 99.90% 10747.90 99.99% 18350.08 max 26214.40 count 160000 ``` batching: ``` ************** RUN 1 EndToEnd Throughput: 11202 ops/sec ************** EndToEnd SERVICE-TIME Latencies distribution in MICROSECONDS mean 1446.31 min 111.10 50.00% 1376.26 90.00% 1482.75 99.00% 3784.70 99.90% 8224.77 99.99% 14024.70 max 17563.65 count 160000 ``` Same scenario: 16 producer / consumers sending to 16 queues 100 bytes persistent msg with replication vs a broker that's already paging due to ``` ./artemis producer --message-count 22000 --message-size 100000 ``` During the whole load test the broker is paging over the 16 queues != TEST and batching still seems beneficial -- 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. For queries about this service, please contact Infrastructure at: [email protected]
