EE Lookup Stage VS Server Transformer

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
Nick_6789
Participant
Posts: 32
Joined: Thu Jan 05, 2006 2:07 am

EE Lookup Stage VS Server Transformer

Post by Nick_6789 »

Hi Guys! Just a lil' curious.

Currently, my project involves utilizing server jobs.

The design is like this. I have a sequential file (main flow),

columns:

1) Customer ID
2) Name

and a lookup hash file (Which serves as independent lookup)

columns:

1) Area ID
2) Address

Utilizing a transformer to do output assignment, my hash file serves as an independent lookup, whereby it does not do any sort of key join with the main flow.

In the hashfile lookup flow, by the 'Address' column, I pre-input the value '56, Washington Street' in the Key Expression column.

What happens is, Area ID (Only related to 56 Washington Street), will be outputted into the output flow. Which means all records from the sequential file will eventually contain the Area ID pertaining to 56 Washington Street.

Can I perform the exact same independent lookup like the above with the 'Lookup Stage' in a parallel job? Or is lookup stage's function is to perform dependent lookups with main flows?

Please advice. Many thanks! =)[/img]
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Change your mindset. Parallel job philosophy is "one stage, one task". The Lookup stage performs lookups. It has no other function. If you want to pre-set the lookup key, do so in an upstream Column Generator stage.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Nick_6789
Participant
Posts: 32
Joined: Thu Jan 05, 2006 2:07 am

Post by Nick_6789 »

ray.wurlod wrote:Change your mindset. Parallel job philosophy is "one stage, one task". The Lookup stage performs lookups. It has no other function. If you want to pre-set the lookup key, do so in an upstream Colu ...
Hi Ray, Thanks for the feedback.

The reason asking this is because I am required to translate the server job design to parallel. I guess, that being too dependent on Server's limited stage availability, and coming to terms of fully utilizing one stage (particularly the transformer stage) to perform multiple tasks, I guess it takes getting used to to the parallel design.

Thanks for the advice though! =D
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Enrol on the Server to Parallel Transition Lab offered by DCS Strategies (the company that funds this site).
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply