Can I process inbound IDocs in parallel to accelerate processing?

Can I process inbound IDocs in parallel to accelerate processing?

We have huge orders coming in, and the job running RBDAPP01 is taking up to 8 hours!

    Requires Free Membership to View

    When you register, you will start receiving targeted emails from my award-winning team of editorial writers. Our goal is to keep you informed on the hottest topics and biggest challenges faced by SAP professionals today.

    Hannah Smalltree, Editorial Director

    By submitting your registration information to SearchSAP.com you agree to receive email communications from TechTarget and TechTarget partners. We encourage you to read our Privacy Policy which contains important disclosures about how we collect and use your registration and other information. If you reside outside of the United States, by submitting this registration information you consent to having your personal data transferred to and processed in the United States. Your use of SearchSAP.com is governed by our Terms of Use. You may contact us at webmaster@TechTarget.com.

Your problem is application performance, not IDoc performance!

I see that you apparently have already decided to process the IDocs in a batch run using RBDAPP01. From there on you are already beyond the interface processing and are within the application, i.e. everything that happens now is due to the processing application, in your case the SAPMV45A application. Parallel processing the IDocs is possible just by planning several RBDAPP01 simultaneous jobs. You would need an intelligent mechanism that guarantees that each jobs picks its own pile of IDocs to process. However, this will not solve your performance problem unless you have several app servers and have every job running on a separate box. There may be locking problems if the IDocs share same material master , customer master data etc, so this may become a difficult enterprise depending on the nature of your data.

I suggest you first try to analyze the IDoc processing performance to find out where the time is lost. This can be done by executing transaction SE30 or the program RBDAPP01 and a single IDoc. Then analyze the result and judge if there are sections that have a long processing time. Have a special look on user-exits and unexpectedly long database selects.

If this does not really help, there is still the old advice from Mama Blue: "Buy a bigger machine!"

This was first published in May 2003