SAP BW Pack performance

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
dsetlteam
Premium Member
Premium Member
Posts: 35
Joined: Mon Feb 10, 2014 10:14 pm
Location: USA

Post by dsetlteam »

I had a bad experience in extracting data from SAP BW Open Hub 4.3.2.1.
Let me share my experience with you first :)

It started with spending lots of time just setting up the open hubs,process chains and source system.Once the connection was setup properly, the next problem arose on extracting data from SAP BW, for some jobs it was fast enough but for others they just kept running even though the process chain at the SAP end had completed in seconds!

The worst nightmare was when the jobs were deployed in production and the jobs completed successfully - but they didn't extract any records (0 records were extracted) even though SAP process chain logs said that they fed many records! This was seen on not on single job but on 5-8 jobs!

Apparently, PMR was raised with IBM and they suggested to use 3 environment variables in each job to rectify the issue.

I don't exactly remember the variable names , but they were something like :-

APT_PROCESS_CHAIN_TIMEOUT
APT_OPEN_HUB_TIMEOUT

In your case, you can check with IBM if there are any environment variables available to speed up the process of data extraction.

Cheers!
asorrell
Posts: 1707
Joined: Fri Apr 04, 2003 2:00 pm
Location: Colleyville, Texas

Post by asorrell »

You should also try increasing your packet size so you send more records at once.
Andy Sorrell
Certified DataStage Consultant
IBM Analytics Champion 2009 - 2020
ggarze
Premium Member
Premium Member
Posts: 78
Joined: Tue Oct 11, 2005 9:37 am

Post by ggarze »

thought about that but in the IBM documentation http://www-01.ibm.com/support/docview.w ... wg21511947 it says that we should use whatever BW has set up.
Post Reply