[
http://jira.jboss.com/jira/browse/JGRP-492?page=all ]
Vladimir Blagojevic updated JGRP-492:
-------------------------------------
We repeated the test that Brian Stansberry uses: 6 node cluster in Atlanta; each node runs
one instance of JBoss; clients were run on remaining machines on the same LAN. We tried
both JGroups 2.5 tcp and udp concurrent stacks. We got the following results:
6 node 4.2GA cluster, tcp conccurent stack, 3 clients each having 1000 threads
4662 req/sec
6 node 4.2GA cluster, tcp conccurent stack, 2 clients each having 1000 threads
5255 req/sec
6 node 4.2GA cluster, tcp conccurent stack, 1 client having 1000 threads
varied between 4432 req/sec and 5929 req/sec
6 node 4.2GA cluster, udp concurrent stack, 3 clients each having 1000 threads
2496 req/sec
6 node 4.2GA cluster, udp concurrent stack, 2 clients each having 1000 threads
2484 req/sec
6 node 4.2GA cluster, udp concurrent stack, 1 client having 1000 threads
varied between 2470 req/sec and 2981 req/sec
Benchmark against JBento ATL cluster
------------------------------------
Key: JGRP-492
URL:
http://jira.jboss.com/jira/browse/JGRP-492
Project: JGroups
Issue Type: Task
Reporter: Bela Ban
Assigned To: Vladimir Blagojevic
Fix For: 2.5
Run the same session replication benchmarks as for 2.4.1 SP3 against 2.5
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira