[ https://issues.jboss.org/browse/JGRP-1722?page=com.atlassian.jira.plugin.... ] Bela Ban commented on JGRP-1722: -------------------------------- Somewhat disappointing, but it looks as if the perf of ENCRYPT is largely a function of the speed of encryption and decryption. Adding a cipher pool didn't really improve performance for regular messages, but should improve it for {{OOB | DONT_BUNDLE}} messages. Parallelization (e.g. concurrent decryption of all messages of a message batch) didn't help, either. The reason is that introducing yet another thread pool probably cancelled out the gains made, through increased context switching. I'll revisit this when baselining JGroups on JDK 7 where a shared fork-join thread pool can be used. Also, uisng the asynchronous invocation API of JGroups would help as the encryption of the response would not have to be done on the same thread as the request, speeding of processing of message batches. I'm resolving this for now, but I'm having a chat with Tristan to see if we can use a faster cipher... -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira