We have been load testing our system lately and we have been noticing many threads waiting to lock the readOwnerList in the LockMap class. The readOwnerList_ is a CopyOnWriteArraySet and (as the sourcecode comment above clearly states) is not the most effecient implementation to use here. So I decided to try and replace the CopyOnWriteArraySet with a ConcurrentHashMap implementation instead and I got a very significant performance boost. Not only was CPU significantly lower on the machines running the cache, but concurrency was also much, much better. I was actually surprised to see how much performance was gained. Therefore, I would certainly recommend that you swap out the CopyOnWriteArraySet to something more efficient. The actual coding took me about 1 minute to do, so it should certainly be a simple way to improve the cache throughput. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4099664#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...