Hi Kiru, We are performing some JMeter Test Plans over Artemis, Here's our numbers:
AWS server-side - 3 Nodes m4.large -> @ CPUS and 4GB RAM, clustered through JGroups + TCPGOSSIP as discovery server - AWS small-type machine with JMeter as client You can find our base Test Plan and the instructions on how to load it over this link: https://github.com/apache/activemq-artemis/tree/master/examples/perf/jmeter For now we've changed this base test plan to support JGroups at the jndi.properties side and currently we are making ~ 15.500 requests per second under a pub / sub sampler, without any JVM tuning yet. The average request is doing around 1ms for low current users and 5ms at 15.500 requests per second. I'm finishing my tests here and I will update the JMeter files with new load / stress / performance tests scenarios. As soon I as have it finished I will submit a PR to activemq-artemis project with those new JMeter files. Hope this helps, Cheers, Glaucio Melo IT Specialist at Movile Brazil -- View this message in context: http://activemq.2283324.n4.nabble.com/Artermis-user-feedback-tp4702928p4702961.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
