Originally posted by: SystemAdmin

Are there any significant differences between using a PUT fundtion versus using an output card to add a message to a queue? Performance-wise? Error messages from failure to write to output card vs. VALID(PUT("MQS", ...), FAIL(LASTERRORCODE()+ ": " + LASTERRORMSG()))? Obviously, you have more control over the construction of the adapter parameters using the PUT function. Any other differences?

Thanks,
Jim Divoky
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: SystemAdmin

The major difference is the scope of transaction - PUT will always have a scope of burst, but with output card you have more flexibility.

But for MQ, I think that's it.
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender
#DataExchange

Originally posted by: SystemAdmin

You have more control concerning error processing in a rule than in an output card. For some adapters this is absolutely necessary because many errors still allow the card and map to complete successfully. For MQ, do not think this extra error control matters.

Yvonne Ricard
Resistance is NOT futile.
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

it depends, if i was going to a mainframe queue manager, put statement in the rule is the best, but just an average que within the same platform, unless your doing something "Fancy" the output card is the way to go.. Are your settings in the IFD for "Client" or "Server" for the output card?
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: SystemAdmin

>> Are your settings in the IFD for "Client" or "Server" for the output card?

No, actually it appears most maps call a queue posting map which is passed the payload and some MQ IDs. I'm not certain why someone thought that was good idea especially when tens or hundreds of thousands of transactions are involved.

Why is "the output card the way to go"? I feel that way also but I can't really give a good argument other than it is more straightforward and in keeping with what I see as good middleware design.

Jim Divoky
#DataExchange
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender