CDC + Split Insert/update

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
algfr
Participant
Posts: 106
Joined: Fri Sep 09, 2005 7:42 am

CDC + Split Insert/update

Post by algfr »

Hello guys,

I have a few questions regarding the CDC stage ?

1) As Suggested by Ray and others, it's better to split inserts and updates. So I use a switch stage after my CDC on the change code and redirect my data to two oracle stage to the same table.
Is this okay to use Load/Append for inserts and Update only for updates ?

2) How do we make sure there is no confilct between inserts and updates ?

3) When I use Load/Append, I get warnings with the dest field that is a ustring. Eveni if I am 100% sure it is a ustring, I get a warning stating I get a ustring to string conversion. If I use Upsert, this does not happen.

Thanks a lot.

Arnaud
algfr
Participant
Posts: 106
Joined: Fri Sep 09, 2005 7:42 am

Post by algfr »

ora_DWH_DIM_PRD_LOC_insert: When checking operator: When binding input interface field "PRD_LOC_NAT_COD" to field "PRD_LOC_NAT_COD": Implicit conversion from source type "ustring[max=30]" to result type "string[max=30]":

--> In load/Append mode only (everything is set to Unicode already)
DSguru2B
Charter Member
Charter Member
Posts: 6854
Joined: Wed Feb 09, 2005 3:44 pm
Location: Houston, TX

Post by DSguru2B »

After splitting the inserts and updates, stage them in seperate datasets and use seperate jobs to insert and update.
Creativity is allowing yourself to make mistakes. Art is knowing which ones to keep.
algfr
Participant
Posts: 106
Joined: Fri Sep 09, 2005 7:42 am

Post by algfr »

Is it really compulsory to create so many jobs ?
Post Reply