Page 1 of 1

Copy tables from one Database to another Database

Posted: Wed Nov 08, 2006 3:07 pm
by shilpa79
Is there something in Datastage that can copy objects(Tables) from one database to another. If there is change in the table (example , addition of a column) in the source database , there should not be any change in the Datastage job as the whole table gets copied.

Re: Copy tables from Database to another Database

Posted: Wed Nov 08, 2006 3:08 pm
by DeepakCorning
Not aware of any such things. May be you can use Database sync utilities to do such activities and not use a ETL tool to do such things. :)

Posted: Wed Nov 08, 2006 3:16 pm
by narasimha
I dont think there is a direct method of doing it in datastage
But if you really want to use datastage for it, there is a way

You can write a stored procedure to move tables from one database to another.
Call this SP from datastage! :wink:

Posted: Wed Nov 08, 2006 4:24 pm
by ray.wurlod
It's called Runtime Column Propagation and you get it in parallel jobs only.

Posted: Wed Nov 08, 2006 10:19 pm
by loveojha2
ray.wurlod wrote:It's called Runtime Column Propagation and you get it in parallel jobs. ...
Runtime Column Propagation, meaning it happens at the runtime only or there are methods of doing it with the job metadata also. :?

Posted: Wed Nov 08, 2006 11:40 pm
by chulett
loveojha2 wrote:Runtime Column Propagation, meaning it happens at the runtime only or...
Hence the name.

Posted: Thu Nov 09, 2006 12:13 am
by tagnihotri
Yes there are ways of doing it with job metadata which is called RCP (Runtime Column Propagation) :wink:
loveojha2 wrote:
ray.wurlod wrote:It's called Runtime Column Propagation and you get it in parallel jobs. ...
Runtime Column Propagation, meaning it happens at the runtime only or there are methods of doing it with the job metadata also. :?

Posted: Thu Nov 09, 2006 12:26 pm
by shilpa79
Runtime Column Propagation is valid ONLY AFTER YOUR INITIAL TABLE STRUCTURE IS DEFINED. that means the table needs to exist on the target database and in case of any additional columns that need to be added can be done using RCP..same with server also you can use a RTI I map creation specifics to do the same functionality.

Posted: Thu Nov 09, 2006 1:01 pm
by ray.wurlod
Not true. If you use direct Read and Write methods, RCP will cause entire rows (all columns) to be propagated even in the absence of a schema file. That's one of the reasons it can be so dangerous! What if the rows have 5000 columns?!!