| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Mapping node vs Compute node |
« View previous topic :: View next topic » |
| Author |
Message
|
| kumar.MBMQ |
 Posted: Thu Oct 10, 2013 3:28 am Post subject: Mapping node vs Compute node |
 |
|
Voyager
Joined: 06 May 2011
Posts: 93
|
Is it better to use mapping node or compute node in message flow.
Message Broker version 8.0.0.2. |
|
| Back to top |
|
 |
| lancelotlinc |
| Posted: Thu Oct 10, 2013 3:29 am Post subject: |
|
|
Jedi Knight
Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA
|
Mapping node for simple maps. Compute node for more complex maps or robust application logic.
Mapping node performs at 35% to 50% of equivalent Compute nodes. Check out the performance reports for your platform. If performance is important to you, you will select Compute nodes every time.
Graphical mapping node is a convenience to the programmer. The cost of that convenience (or the price for using a Mapping node) is slower performance.
| Quote: |
| V8 introduces a new Mapping node which allows the user to visually map and transform data from source to target. This new mapping node has excellent performance characteristics, and is a viable option for performance sensitive transformations. Some tests have been measured performing close to optimised programmatic transformations in ESQL, Java and .Net, with the typical measurement being 50%. |
I would not characterize 50% horsepower as 'excellent'. And by the way, 50% is in the best circumstance, and your mileage may be closer to 20%. These performance measurements and statements are directly from IBM.
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER |
|
| Back to top |
|
|
| mqjeff |
| Posted: Thu Oct 10, 2013 5:51 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
The usual performance issue people have with the mapping node is that it is compiled at first use, which can take quite a while.
So the first message through can take 5 minutes, where the next message takes .1 second.
The v8 mapping node compiles down to Java byte code, so it's not any slower to execute, once compiled, than a JCN.
But the mapping node gets unwieldy as the complexity of the mapping grows. And the time to develop it does to. Just wiring in a straight forward map of 100 fields can take significantly longer than cut/pasting ESQL set statements.
On the other hand, the mapping node can autogenerate map relations, and you can't autogenerate ESQL set statements.
If you notice that I've not said "use one or the other" it's because there isn't an answer. |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
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
|
|
|
|
