I have executed and APD thru Process chain.
The job for APD : BI_PROCESS_ANPR is running more than 30 hrs.
In my case APD is transfering data from BI DSO to CRM table.
DSO has around 1.5 million data and 160 fields.
Please anyone can give an idea or share doc,to analysis why it is taking long time.
How do I parrallize the job or any other way to improve preformance.
Help will be appreciated.
Thanks for your reply,
But since 30 hrs in job log i can see only :
Step 001 started (program RSPROCESS, variant &0000000000165, user ID BWREMOTE)
Start process ANPR ZADS_CRM in run XXXXXXXX of chain XXXXXX
And in target it is showing 0 records.
Even I am cheking, is their any way we can check how many records are updated. Because the commit statement is called only after all records are updated.
Any other suggestions or help?
I guess, you are transferring data into a CRM system. This might be a problem depending on what is triggered on CRM side (to be analyzed). Hence, I would assume that APD is waiting most of the time for CRM sytem to get things done.
If possible, do a search for notes on CRM as target, I think performance related ones exist.
Parallelization needs to split data on business partner level. If you model a "template" analysis process with an apppropriate filter, you can create copies of it and adapt the filter in each one (e.g. intervals for business partner number, or other values of BP attributes). Those copies could run in parallel.
However, I would first check the activities happening in CRM (lots of subsequent updates for each BP maybe?).