Skip to Content

These days concept of “parallel processing “is very popular between SAP techies.  Parallel processing is good, but less knowledge and lack of communication some time create a serious problem in production environment.

End of the day it become checkmaet game between Basis and ABPer with long list of counter arguments:

Basis

Developer

CPU utilization is high not sure what this program is doing, occupying all the available process.

My program is working fine, server don’t have resource.

Something wrong in program can you please debug the program.

I am using standard SAP function module. I don’t know what is happening.

Database transaction log is full. You are not passing commit in your program.

I did that, Looks like db issue.

Have you tested your program in QA

Yes we did no issue in QA.

 

Some time it require effort and cost to fix such problems.  We can avoid this by implementing simple quadrant rule from the beginning of project:

image

 

Socialization: Many times during sizing we never consider parallel processing.  This is responsibility of development team to socialize PMO and relevant teams (basis & functional) about number of programs in which they are planning to use parallel processing, detail description and documentation of technique planning to use for parallel processing  and their impact like

·         Parallel processing using background jobs

·         Parallel processing using synchronized RFC.

Infrastructure for Testing:  Basis is responsibility for providing infrastructure for testing a parallel processing. A dedicated basis resource is required to analyze the infrastructure requirement for parallel processing. Keep in mind that due to less data parallel processing works fine some times in QA environment.

Configuration:   Based on the information provided by development team basis is responsible for configuring following:

 

1.       Background server group

2.       Sufficient number of update processes

3.       Sufficient space for transaction log files

4.       RFC server group and relevant parameters like :

rdisp/rfc_check

rdisp/rfc_use_quotas

rdisp/rfc_max_queue

rdisp/rfc_max_login

Scheduling:   A proper scheduling is very important part of parallel processing, while scheduling a parallel processing job following things need to consider:

1.       Available server resources

2.       Schedule conflict with other parallel processing job

3.       Avoid scheduling job while online backup is running

4.       Avoid scheduling during finance month end process

To report this post you need to login first.

Be the first to leave a comment

You must be Logged on to comment or reply to a post.

Leave a Reply