InterProcess stage

[EJRoufs]

I am using DataStage version 7.1.

I am starting to use the InterProcess stage for the first time, and am looking for any helpful advice in its usage.

I do have some general questions concerning it. For example, is this stage at all beneficial when taking 4 files and using a link collector and merging them into 1 file? I was thinking it wouldn't be helpful here, but when putting 1 in, I do see slight improvement. I didn't know if it was just coincidence (didn't run it hundreds of times), or if it does actually help here. In some minimal training I had on this stage, I was thinking it would be most helpful where Transform stages are used.

Thanks for any advice. :>

[Sainath.Srinivasan]

It will be beneficial if you have more than one processor in your system.

[EJRoufs]

It will be beneficial if you have more than one processor in your system.

Yep, Windows Server with 8 processors.

[rajkraj]

I am using DataStage version 7.1.

I am starting to use the InterProcess stage for the first time, and am looking for any helpful advice in its usage.

I do have some general questions concerning it. For example, is this stage at all beneficial when taking 4 files and using a link collector and merging them into 1 file? I was thinking it wouldn't be helpful here, but when putting 1 in, I do see slight improvement. I didn't know if it was just coincidence (didn't run it hundreds of times), or if it does actually help here. In some minimal training I had on this stage, I was thinking it would be most helpful where Transform stages are used.

Thanks for any advice. :>

Link collector is very help full and it will increase the process speed in server jobs also....

[ray.wurlod]

Read Chapter 2 of the Parallel Job Developer's Guide (parjdev.pdf), which explains pipeline and partition parallelism.
The IPC stage (or row buffering) in server jobs achieves pipeline parallelism.
Multi-instance running, Link Partitioner stage and multiple streams within a job are ways to implement partition parallelism in server jobs.