Performance Reports for MQ 7.5 - where art thou?

[Potkay, Peter M (CTO Architecture + Engineering)]

This Redbook talks about the performance gains in MQ 7.1 compared to MQ 6 and MQ 7.0.
http://www.redbooks.ibm.com/redbooks/pdfs/sg248087.pdf


This page talks about some significant performance gains on top of that in MQ 7.5 compared to 7.1.
http://www-01.ibm.com/software/websphere/subscriptionandsupport/compare-mq-versions.html


Yet the Performance Reports page still doesn't have reports for MQ 7.5. Any word on when they might show up? Seems like with those performance improvements IBM would be eager to highlight them.
http://www-01.ibm.com/support/docview.wss?uid=swg27007150



Peter Potkay




************************************************************
This communication, including attachments, is for the exclusive use of addressee and may contain proprietary, confidential and/or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you are not the intended recipient, please notify the sender immediately by return e-mail, delete this communication and destroy all copies.
************************************************************

To unsubscribe, write to LISTSERV-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org and,
in the message body (not the subject), write: SIGNOFF MQSERIES
Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://listserv.meduniwien.ac.at/archives/mqser-l.html

[Andy Hickson]

There has been a continuing Hursley focus on queue manager performance for
many years, and while this continues in MQ V7.5 the performance differences
between 7.1 and 7.5 are generally relatively minor. The numbers quoted in
the referenced page look more like the enhancements between 7.1 and 7.0.1,
as measured by the Hursley benchmarks.

There are a few specific 7.5 sparklers, for example 7.5 adds support for
multiple cluster transmit queues (in some environments this might provide
very significant performance advantages) but in more general terms 7.5 is
expected to have a relatively similar performance profile to 7.1.

While on the subject of the Hursley performance benchmarks (and at the risk
of opening the flood gates) I would be interested in customer observations
on their perception of the relative performance of 7.0/7.1/7.5 and in any
particular performance pain points.
I would also be interested in hearing to sort of message rates that you are
actually achieving in production, for example MP7L measures client
applications driving 2KB non persistent messages at 30,000 msg/sec and 2KB
persistent at 10,000 msg/sec, for a request/reply application on relatively
modest servers, how does this compare with message rates being achieved in
production scenarios ?

To unsubscribe, write to LISTSERV-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org and,
in the message body (not the subject), write: SIGNOFF MQSERIES

[Potkay, Peter M (CTO Architecture + Engineering)]

Andy,
Then that page is very misleading.

http://www-01.ibm.com/software/websphere/subscriptionandsupport/compare-mq-versions.html
"Performance Points of MQ v7.5 vs. MQ v7.1"

It specifically touts big performance gains in MQ 7.5 when compared to 7.1, and it was the deciding factor for us going with 7.5 over 7.1. (At that decision point 7.1 already had a couple of fixpacks out - 7.5 did not).

There isn't any Feedback link on that page. Can you bring this to the attention of the Product Manager or someone that can get that typo fixed? It should read "Performance Points of MQ v7.5 vs. MQ v7.0.1"?

Thanks for the clarification, it means the MQ 7.1 Performance Reports have the info I need for my 7.5 set ups.


Peter Potkay

-----Original Message-----
From: MQSeries List [mailto:MQSERIES-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org] On Behalf Of Andy Hickson
Sent: Friday, July 26, 2013 8:57 AM
To: MQSERIES-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org
Subject: Re: Performance Reports for MQ 7.5 - where art thou?

There has been a continuing Hursley focus on queue manager performance for many years, and while this continues in MQ V7.5 the performance differences between 7.1 and 7.5 are generally relatively minor. The numbers quoted in the referenced page look more like the enhancements between 7.1 and 7.0.1, as measured by the Hursley benchmarks.

There are a few specific 7.5 sparklers, for example 7.5 adds support for multiple cluster transmit queues (in some environments this might provide very significant performance advantages) but in more general terms 7.5 is expected to have a relatively similar performance profile to 7.1.

While on the subject of the Hursley performance benchmarks (and at the risk of opening the flood gates) I would be interested in customer observations on their perception of the relative performance of 7.0/7.1/7.5 and in any particular performance pain points.
I would also be interested in hearing to sort of message rates that you are actually achieving in production, for example MP7L measures client applications driving 2KB non persistent messages at 30,000 msg/sec and 2KB persistent at 10,000 msg/sec, for a request/reply application on relatively modest servers, how does this compare with message rates being achieved in production scenarios ?

To unsubscribe, write to LISTSERV-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org and, in the message body (not the subject), write: SIGNOFF MQSERIES Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://listserv.meduniwien.ac.at/archives/mqser-l.html
************************************************************
This communication, including attachments, is for the exclusive use of addressee and may contain proprietary, confidential and/or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you are not the intended recipient, please notify the sender immediately by return e-mail, delete this communication and destroy all copies.
************************************************************

To unsubscribe, write to LISTSERV-0lvw86wZMd9k/bWDasg6f+***@public.gmane.org and,
in the message body (not the subject), write: SIGNOFF MQSERIES