qpid-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marnie McCormack <marnie.mccorm...@googlemail.com>
Subject Re: Performance troubles with qpid and mule
Date Tue, 24 Nov 2009 10:34:23 GMT
Hi Henry,

Mule is pretty unhelpful in the way that it handles publication - in that it
creates a connection (a session, a producer) for each send()and discards it
- starting again for the next message. So it's pretty slow. I'm not so sure
why consumption would be slow though - for transient messages you should be
getting performance from Qpid orders of magnitude better.

I know some other users of Mule with Qpid so I'll go and find out how they
got performance to be better and come back to you shortly. Could be
transaction related ?

As an aside, Mule can introduce challenges from a dev perspective i.e. the
use of Object messages etc. It possibly does always make things slower (and
harder to tune), but I guess the trade off is config/flexibility.

On Tue, Nov 24, 2009 at 1:18 AM, Henry Robinson <henry@cloudera.com> wrote:

> Hi -
> I'm getting disappointing performance reading from a Qpid queue into a
> simple Mule endpoint (that does essentially nothing) - about 150 msgs/s.
> I'd
> appreciate some guidance about performance tuning.
> This is all running on the same machine - an 8 core Xeon machine with 8Gb
> of
> RAM. Here's the relevant JMS configuration in Mule:
> <jms:connector name="JMSConnector"
>                   connectionFactory-ref="qpidfactory"
>                   specification="1.1"
>                   persistentDelivery="false"
>                   numberOfConsumers="10"
>                   maxRedelivery="2">
>    <receiver-threading-profile doThreading="true" maxThreadsActive="10"
> maxThreadsIdle="5" threadTTL="60000"/>
>    <dispatcher-threading-profile doThreading="true" maxThreadsActive="10"
> maxThreadsIdle="5" threadTTL="60000"/>
>  </jms:connector>
> If I jstack the mule process (unscientific, to be sure, but attaching a
> profiler causes the whole thing to die slowly under timeouts) I notice that
> the Dispatcher threads are almost all blocked on
> AMQConnection.createSession
> (which has a single-occupant critical section). So should I have enabled
> something that allows for session reuse? Alternatively, are there some
> other
> performance tweaks I can attempt? (I have looked here:
> http://qpid.apache.org/how-to-tune-m3-java-broker-performance.htmlalready).
> Sorry if this is actually a Mule question, but as you can see the issue
> appears to be in Qpid code. Thanks in advance for your advice.
> cheers,
> Henry

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message