Improving the performances of JMS-based applications

In the Java world, a standardised interface exists for Message-Oriented Middleware (MOM): Java Messaging Service or JMS. Like other middleware, some JMS implementations use clustering techniques to provide some level of performance and fault tolerance. In this paper, we analyse the efficiency of various clustering policies in a real-life cluster and the key parameters impacting the performances of MOMs. We show that the resource efficiency of the clustering methods can be very poor due to local instabilities and/or global load variations. To solve these issues, we describe the rules that control these parameters for optimal performances and propose a solution based on autonomic computing to (1) dynamically adapt the load distribution among the servers (load-balancing aspect) and (2) dynamically adapt the replication level (provisioning aspect). We present an evaluation that shows the impact of these rules on the performances and behaviour of the dynamic provisioning of replicated queues.