MQSeries.net :: View topic - Run into WMB Parser Bottleneck

MQSeries.net

Tech Exchange

Education

Certifications

Library

Info Center

SupportPacs

FAQÂ Â

Usergroups

RSS Feed - WebSphere MQ Support

RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Run into WMB Parser Bottleneck

Run into WMB Parser Bottleneck

« View previous topic :: View next topic »

Author

Message

poonchikit

Posted: Sun Aug 12, 2007 11:24 pm Post subject: Run into WMB Parser Bottleneck

Novice

Joined: 04 Oct 2006
Posts: 10

My company uses WMB to transform XML to BLOB messages and the other way round between open and mainframe applications.

In the current exercise, my company changed the XML message format on the front-end application to support txn format changes in the mainframe systems. After that, we found the CPU consumption rate shot up to 100%. We consulted IBM and they said we ran into built in XML parser or WMB inate limitations. Worse yet, when we lengthened the XML test message successively, we found that the CPU usage rate increases exponentially. All messages were less than 4K long.

We use the same mechanism and algorithm before and after the XML format change. The top management is considering to dump WMB if no solution can be found. This is ridiculous to see enterprise software to exhibit this non-linear scalability.

Is there any way to replace WMB's built-in XML parser?

Vitor

Posted: Mon Aug 13, 2007 12:24 am Post subject:

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

You don't say which version of WMB & on what platform so this may be more rubbish than normal. If you're not on v6 on a grown up platform then this may be your way forward!

I'll stand to be corrected by the next IBMer who passes but AFAIK WMBv6 uses the Xerces parser under the covers which is pretty good and fairly standard so I'm not sure what you'll replace it with. If it can be replaced and on that I have no information whatsoever.

For what it's worth, we use WMBv6 to transform XML to CWF and back with no particular performance issues. Fairly surprised IBM have come back with such a "sorry you're doomed". Also interested to know why you have BLOB as a format; you might do better with formatted output so you don't need bitstreaming.

They'll be a more authorativite IBMer along in a moment I feel sure. I'm just some guy.

_________________
Honesty is the best policy.
Insanity is the best defence.

kimbert

Posted: Mon Aug 13, 2007 1:12 am Post subject:

Jedi Council

Joined: 29 Jul 2003
Posts: 5542
Location: Southampton

This is almost certainly solvable. Please post
- the version of WMB that you are using
- the place where the CPU time is being consumed ( XML parsing or BLOB output )
- Details of the message flow ( are you using MRM with the CWF physical format? )

I would be suprised if the bottleneck was in XML parsing. I know of a problem with older versions of WMB when outputting large MRM CWF messages. That was fixed in v6.

Display posts from previous:

Page 1 of 1

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Run into WMB Parser Bottleneck

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

Protected by Anti-Spam ACP