Archives of the TeradataForum

Message Posted: Thu, 14 Aug 2003 @ 13:45:31 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Dwight,

I have some follow-up/clarification questions if you wouldn't mind.

You mention "workflow", so I'm assuming that you are on 6.2 of Informatica.

It sounds like you have unchecked the "load" option on the target to get this result. Why have you opted to perform the mload as a separate process rather than just loading automatically with Informatica? Is this for re-startability only?

In some cases, we may run into TABLEWAIT/buffer issues when streaming data from a database, rather than a file. Have you experienced a problem with this at all. Informatica implies that locks could be maintained on the source tables for quite some time. Separating the extraction from the load (as it appears you have done) would remedy this as well, I suppose. I guess that's another advantage of splitting up the creation of the data file and the performing the mload to the target table, eh?

I also understood your note to imply that there are cases where you are executing sql scripts rather than relying on Informatica for all transformations. Can you provide any type of general example of when you have found that to be the case. I'm guess I'm wondering what the "driver" was in these cases, such as data volume, or transformation complexity, etc.

Thanks a lot for the feedback. I really appreciate you taking the time to provide input on this.

Duane

	<Prev	Next>		<<First	<Prev	Next>	Last>>