> The throughput of the benchmarking program on NT workstation with
> 1.2c
> was about 5 msgs/sec. I tried it on a solaris (Ultra 5) machine and the
> performance was less than that of NT workstation. I am using
> point-to-point
> messaging model. The queue, sender and receiver benchmarking programs
> are
> all on the same machine. I am not sure if there's some configuration
> issue
> that's causing the poor performance. Can you shed some light on this?
Actually not. We are investigating this behaviour. Interestingly, it
appears on *some* NT WS, not on all, also on some AIX/Solaris, not on all.
The behaviour is dependent on the message size. If the actual size is >
2048, it appears, otherwise not. The 1.2c has default compression with a
threshold of 2KB. How large are your message bodies? Our first thoughts
are the TCP buffers, but they are 8192; same as on a NT Server where it
runs fine. We hope to resolve this strange behaviour for the next release.
--
Andreas Mueller, [EMAIL PROTECTED], IIT GmbH, Bremen/Germany, http://www.iit.de
SwiftMQ, JMS Enterprise Messaging System, http://www.swiftmq.com
------------------------------------------------------
SwiftMQ developers mailing list * http://www.swiftmq.com
To unsubscribe from this list, send an eMail to
[EMAIL PROTECTED] and write in the body of your message:
UNSUBSCRIBE developers <your-email-address>
Archive: http://www.mail-archive.com/developers@mail.iit.de/
