[
https://issues.jboss.org/browse/ISPN-6494?page=com.atlassian.jira.plugin....
]
Johno Crawford commented on ISPN-6494:
--------------------------------------
[~belaban] any updates on this? The default infinispan configurations still reference
no-bundler which we currently have configured for our application.
Our application is a little special in the sense that we hijack a fork channel from the
infinispan jgroups stack and use that for internal cluster communication and infinispan
for various distributed / replicated caches.
I have observed that the message rate is so high that the executor starts rejecting tasks
and creating fire and forget threads which really hurts performance.
We would like to investigate bundlers / batching as opposed to increasing thread count to
see if that resolves our performance issues.
Investigate bundler performance
-------------------------------
Key: ISPN-6494
URL:
https://issues.jboss.org/browse/ISPN-6494
Project: Infinispan
Issue Type: Task
Components: Core
Affects Versions: 9.0.0.Alpha1
Reporter: Dan Berindei
Assignee: Dan Berindei
Priority: Major
Fix For: 10.0.0.Final
For ISPN-6027 we changed the default JGroups bundler to {{sender-sends-with-timer}},
because it was faster in some of the performance tests. However, IspnPerfTest shows
{{transfer-queue-bundler}} to be consistently better, so we need to investigate the
bundler choice again.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)