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 DataPower » workload management question

Post new topic  Reply to topic Goto page Previous  1, 2
 workload management question « View previous topic :: View next topic » 
Author Message
anveshita
PostPosted: Thu Apr 28, 2011 9:24 am    Post subject: Reply with quote

Master

Joined: 27 Sep 2004
Posts: 254
Location: Jambudweepam

Vitor wrote:

What proportion of small messages to large do you have? What is the difference in SLA?


80% of the messages are small 20% Large. Small gets processed in 0.6 sec and large in 5-6 sec

Quote:


- How did you arrive at 4 queues? Is that just for illustrative purposes in this thread?


Yes it is just for an example
Quote:


- Why are there the same number of queues for large and small messages? As you talked about timeouts. I'm assuming here the small messages have a short SLA? Why do you not have 3 queues (using this example) working the short messages & 1 chewing through the long?


Yes three(Q1,Q2,Q3) can be grouped and 1 queue say Q4 can be left for large messages. Now if Q4 is empty will it be possible to route the small messages to Q4? If so what is a better way if there is any,to do this?
Quote:


- Why is there only 1 processor per queue? Is this a restriction of the package or just how you've written this example? It's unusual for any application, even a package, to open a queue exclusively.

Consider the package as a blackbox and the requirement was to have one processor per queue. Yes I agree it is unusual, but have to live with it for now.
Quote:

- Why are you so keen to avoid idle queues (by routing different message types)?


Just so I can make all consuming processes engaged.

Thanks for you input.
Back to top
View user's profile Send private message
Vitor
PostPosted: Thu Apr 28, 2011 9:48 am    Post subject: Reply with quote

Grand High Poobah

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

anveshita wrote:
Quote:

- Why are you so keen to avoid idle queues (by routing different message types)?


Just so I can make all consuming processes engaged.


I don't see that as an especially compelling requirement. I've never seen any site that doesn't have (and doesn't have a problem with) applications which occassionally idle while looking for work.

I certainly don't think it's worth tying yourself in knots, building a convoluted solution with multiple failure points that the support staff will never properly understand and risking downtime just to be able to say "I'm using all the available application instances".

(On most modern OS they'll get paged out and their resource reallocated anyway).

But that's just what I think. Other opinions are equally valid.

And push back on that 1 processor to a queue. It's odd, and I'd like to know the technical reason. It smells like message affinity to me...
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
anveshita
PostPosted: Thu Apr 28, 2011 9:55 am    Post subject: Reply with quote

Master

Joined: 27 Sep 2004
Posts: 254
Location: Jambudweepam

Quote:
don't see that as an especially compelling requirement. I've never seen any site that doesn't have (and doesn't have a problem with) applications which occassionally idle while looking for work.


Its not a requirement, but rather a wishlist item. I should be able to process the small messages piling in Q1,Q2,Q3 by routing them to Q4 if I find the queue is empty and processor p4 is idle. If I control P4, I could put logic that checks if Q4 is empty then pull messages from Q1/Q2/Q3.

Quote:
And push back on that 1 processor to a queue. It's odd, and I'd like to know the technical reason. It smells like message affinity to me...

Well it could be. I am still trying to find out, but its a vendor product, so I have less chance to get an answer.
Back to top
View user's profile Send private message
Vitor
PostPosted: Thu Apr 28, 2011 10:17 am    Post subject: Reply with quote

Grand High Poobah

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

anveshita wrote:
Its not a requirement, but rather a wishlist item. I should be able to process the small messages piling in Q1,Q2,Q3 by routing them to Q4 if I find the queue is empty and processor p4 is idle. If I control P4, I could put logic that checks if Q4 is empty then pull messages from Q1/Q2/Q3.


That assumes that P1-P4 are running 1 per processor. Which may be, but could also not be, true.

I stand by my assertion that it's easier and more maintainable not to route by queue depth. But you must do what you think is right.
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
anveshita
PostPosted: Thu Apr 28, 2011 10:24 am    Post subject: Reply with quote

Master

Joined: 27 Sep 2004
Posts: 254
Location: Jambudweepam

Quote:
That assumes that P1-P4 are running 1 per processor

Ironically that is the fact p1-P4 are running 1 per processor
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  Reply to topic Goto page Previous  1, 2 Page 2 of 2

MQSeries.net Forum Index » WebSphere DataPower » workload management question
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.