Sort with Join stages

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
devidotcom
Participant
Posts: 247
Joined: Thu Apr 27, 2006 6:38 am
Location: Hyderabad

Sort with Join stages

Post by devidotcom »

Hi All,

I have a scenerio where I sort and hash partition data on column 1 and column 2 also I would have to rename these columns to col 1 and col 2.

The output link then is an input to the join stage which has same partitioning on this link. The other input link to the join would have the same names col 1 and col 2. Will this work? Would it have the same hash partitioning done on the sort stage.

Thanks
Devi
Maveric
Participant
Posts: 388
Joined: Tue Mar 13, 2007 1:28 am

Post by Maveric »

Yes it would propagate the sort order and partitioning. specify the partitioning as SAME in the join stage for this link. Ideally for renaming you should use a copy or Modify or transformer stages. So it would be Sort -> copy or Modify or transformer stages -> Join. And partitioning is SAME for the Rename stage and join stage.
Maveric
Participant
Posts: 388
Joined: Tue Mar 13, 2007 1:28 am

Post by Maveric »

Forgot. If you are using Transformer stage to rename, then it will not propagate the sort order by default. You need to check the property "preserve sort order" on the output tab of transformer.
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

Also make sure the other link of the join is also partitioned and sorted on the same columns.
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'
Post Reply