[jboss-jira] [JBoss JIRA] (AS7-4930) Better JGroups defaults
[
https://issues.jboss.org/browse/AS7-4930?page=com.atlassian.jira.plugin.s...
]
Bela Ban edited comment on AS7-4930 at 5/31/12 11:20 AM:
---------------------------------------------------------
Below is a config I think would make sense. Note that the core and max threads should
probably changed, based on the number of channels you use over the shared transport.
{{
<subsystem xmlns="urn:jboss:domain:threads:1.1">
<bounded-queue-thread-pool name="jgroups"
allow-core-timeout="true">
<core-threads count="4"/>
<queue-length count="20000"/>
<max-threads count="200"/>
<keepalive-time time="10000"
unit="milliseconds"/>
</bounded-queue-thread-pool>
<bounded-queue-thread-pool name="jgroups-oob"
allow-core-timeout="true">
<core-threads count="2"/>
<queue-length count="1000"/>
<max-threads count="20"/>
<keepalive-time time="10000"
unit="milliseconds"/>
</bounded-queue-thread-pool>
<scheduled-thread-pool name="jgroups-timer">
<max-threads count="4"/>
</scheduled-thread-pool>
</subsystem>
}}
was (Author: belaban):
Below is a config I think would make sense. Note that the core and max threads should
probably changed, based on the number of channels you use over the shared transport.
<subsystem xmlns="urn:jboss:domain:threads:1.1">
<bounded-queue-thread-pool name="jgroups"
allow-core-timeout="true">
<core-threads count="4"/>
<queue-length count="20000"/>
<max-threads count="200"/>
<keepalive-time time="10000"
unit="milliseconds"/>
</bounded-queue-thread-pool>
<bounded-queue-thread-pool name="jgroups-oob"
allow-core-timeout="true">
<core-threads count="2"/>
<queue-length count="1000"/>
<max-threads count="20"/>
<keepalive-time time="10000"
unit="milliseconds"/>
</bounded-queue-thread-pool>
<scheduled-thread-pool name="jgroups-timer">
<max-threads count="4"/>
</scheduled-thread-pool>
</subsystem>
Better JGroups defaults
-----------------------
Key: AS7-4930
URL: https://issues.jboss.org/browse/AS7-4930
Project: Application Server 7
Issue Type: Enhancement
Components: Clustering
Affects Versions: 7.1.2.Final (EAP)
Reporter: Tristan Tarrant
Assignee: Paul Ferraro
The JGroups subsystem needs optimal defaults out of the box, that are also good for
derived products (e.g. JDG), both for protocols and thread pools.
Use a bounded thread pool for the default executor
core-size==number of members (32 for JDG)
max-threads = 1.5*core-size
oob should be queueless
AS7 protocol list (UDP): PING, MERGE2, FD_SOCK, FD, VERIFY_SUSPECT, BARRIER,
pbcast.NAKACK, UNICAST2, pbcast.STABLE, pbcast.GMS, UFC, MFC, FRAG2, RSVP
JDG protocol list (UDP): PING, MERGE2, FD_SOCK, FD_ALL, pbcast.NAKACK, UNICAST2,
pbcast.STABLE, pbcast.GMS, UFC, MFC, FRAG2, RSVP
AS7 protocol list (TCP): MPING, MERGE2, FD_SOCK, FD, VERIFY_SUSPECT, BARRIER,
pbcast.NAKACK, UNICAST2, pbcast.STABLE, pbcast.GMS, UFC, MFC, FRAG2, RSVP
JDG protocol list (TCP): MPING, MERGE2, FD_SOCK, FD, VERIFY_SUSPECT, pbcast.NAKACK,
UNICAST2, pbcast.STABLE, pbcast.GMS, UFC, MFC, FRAG2, RSVP
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira