Run time column propagation

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
hhh
Participant
Posts: 86
Joined: Tue Aug 02, 2005 7:39 am

Run time column propagation

Post by hhh »

Hello friends,

What is Run time column propagation check box in Administrator ?

I have read the help like

If you enable this feature, stages in parallel jobs can handle undefined columns that they encounter when the job is run, and propagate these columns through to the rest of the job. This check box enables the feature, to actually use it you need to explicitly select the option on each stage .

But I could not get properly.

Can u please say, what is this features and where is it useful in parallel jobs ?

Thanx in advance

Regards

Hman
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

HI,
You already have lots of thread reg this issue in this forum.
There is disadvantages given for this check (RCP). Metadata wont be registered for the intermediate stages.
Yes if it is checked, it carries away the coloumn parameters such as datatype and defauts to the further stages.
If it is checked, it would reduce some work around. But Experts say better not to enable it.

regards
deepak
bcarlson
Premium Member
Premium Member
Posts: 772
Joined: Fri Oct 01, 2004 3:06 pm
Location: Minnesota

Post by bcarlson »

kumar_s wrote:HI,
But Experts say better not to enable it.
I think that is putting it a bit too strong. It really depends on the project. Our shop does not use the metadata (we have a different tool for that), so RCP does not hinder us in that respect.

RCP is extremely handy when you are dealing with very large records (hundreds of fields) or systems where the record layouts may change frequently.

Another RCP can be very useful is in the case of a complex program. The more complex the program (more stages, that is), the more places you have to correct table definitions. With RCP you change it once at the beginning and the changes will flow through the rest of the program. Field-specific references would obviously have to change as well, but that happens with or without RCP. Yes, I am over-simplifying the task with RCP, but the point is that it really does ease maintenance.

We turn on RCP as the default, right at the project level, and only turn it off when necessary.

But, while I disagree with Deepak on the use of RCP, I do agree that you should do some more research and check out the threads on this topic elsewhere in the formum.

Good luck,

Brad.
Post Reply