On 08/15/2013 02:44 PM, zackhasit wrote:
Thanks. From looking at the profiler it looks like there are a lot more
threads running than before (6 threads per connection to broker) and most
increase might be context switching. Is it possible to reduce threads in
general to bare minimum.
As I see it :
IOTranport thread
DedicatedTaskRunner
CompositeTaskRunner - Not sure ?
3 timer threads
Can these be reduced ? Can you explain if all are necessary and what does
Composite taskruner thread do ?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/High-CPU-on-CMS-application-tp4670425p4670434.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Sounds like you use Failover and those are threads that are used to
manage client failover and reconnection. The task runners will idle
most of the time if you are connected, those only kick in once the
connection dies and the transport needs to reconnect in the background.
The inactivity monitor uses its own timer and task runner for async
heartbeats and read checks on the connection.
You'd need to further profile to see if any of those threads are really
doing much work or not.
--
Tim Bish
Sr Software Engineer | RedHat Inc.
tim.b...@redhat.com | www.fusesource.com | www.redhat.com
skype: tabish121 | twitter: @tabish121
blog: http://timbish.blogspot.com/