I have a question. Do we want to spend time automating those tests so that we can for example leave them running for the night, or is it fine to just make them work to be able to get some results from time to time? The reason I'm asking is that if we want to automate, I think it would be useful if the tests could start the relevant provider, run a test against it, then stop it (and maybe clean up) - this would prevent a failing provider from affecting the others (if a provider consumes a significant amount of RAM, all others are probably going to be slowed down so the comparison would not be fair). Or is not needed at this point and we just want to be able to get some nice graphs and are willing to do some manual work to get them? Also, do we want to spend time finding out why QPid and ActiveMQ don't work with our tests? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4043581#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...