ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » clustered vs. non-clustered

Post new topic  Reply to topic
 clustered vs. non-clustered « View previous topic :: View next topic » 
Author Message
jrannis
PostPosted: Wed Jan 23, 2002 12:09 pm    Post subject: Reply with quote

Novice

Joined: 04 Nov 2001
Posts: 10
Location: Waterloo Ontario

Can someone point me to some information as to why one would NOT want an MQ Clustered environement (in a multi queue manager multi server environment).

Or if someone want to offer their own opionion that would be great!
Back to top
View user's profile Send private message
mpuetz
PostPosted: Wed Jan 23, 2002 4:25 pm    Post subject: Reply with quote

Centurion

Joined: 05 Jul 2001
Posts: 149
Location: IBM/Central WebSphere Services

Hi,

pros:

- currently only way to scale throughput of messages with multiple
cluster qmgrs and brokers
- adminstrative ease of use since you no longer define individual
channels between qmgrs (effort grows like N*N, N being the number
of qmgrs)

cons:

- administration can be unforgiving. Make a mistake during administration
and it can be hard to recover. Never administrate a cluster using
MQExplorer, use scripts only and test them well.

- messages can get trapped on a qmgr if either the qmgrs or its broker fail
until the recovery process restarts the service on a standby machine
(typically takes 5 minutes).
If you have a low latency, persistent message scenario you can get into
trouble here, since messages can exceed the maximum allowed processing
time (if its less than 5 minutes).
Currently the only *true* solution to this problem is using non-persistent
messaging with a request-reply protocol using message expiry. This is harder
to implement than persistent fire-and-forget (but also much faster).
MQ 5.3 introduces persistent shared queues on the 390 platform which allows
you to use persistent fire-and-forget even for those scenarios. Shared queues
are *real* cluster queues in the sense that they don't belong to a particular
qmgr but are external objects which reside in the so-called coupling facility
(basically a big bunch of shared memory) of the IBM 390. If one qmgr fails,
other will still be able to process the messages on the shared queue.
(MQ 5.2 only offers non-persistent shared queues which are not resistant to
coupling facility break-downs, although the latter are very unlikely).


_________________
Mathias Puetz

IBM/Central WebSphere Services
WebSphere Business Integration Specialist
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  Reply to topic Page 1 of 1

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » clustered vs. non-clustered
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
 
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.