Hi, I am facing a trivial situation. I am testing the performance of JBoss with Mule. Here is the scenario:- 1. I have built a Mule application and deployed on jboss server in which I have exposed one http endpoint and passed one parameter. 2. Inbound transformer will extrac the request and convert it to java bean. 3. This java bean is passed to UMO which will not do anything but passes it to response tranformer. 4. This java bean is converted to string and passed as a http-string as a output. The problem is the performance. The result is very inconsistent. Response time varies a lot for same request. I performed this test several times even restarting the jboss server. Please let me know where i am wrong or is it a known pattern for jboss. My boss is really pushing me to sort this out. Immediate responses are highly appreciated. -- View this message in context: http://www.nabble.com/Query-related-to-JBoss-performance-with-MULE-tp1604... Sent from the JBoss - Dev mailing list archive at Nabble.com.