Hi Naveen.
I'm not an expert on memory management but maybe I can add some points:
- Running the same ABAP twice usually ends up with two work processes running the same code (assuming the program is not constantly rolling in/out and therefore both can run on the same work process).
- There are (to my knowledge) no resource limits on a 'per report' level but on a 'per work process' level. Meaning if ZABC get's 10MB and a second ZABC starts it will also get 10MB. But if the server can only provide 15MB in total you will see that one gets 10 and the other only 5 (or both get less than 10)
- I think the SAP help portal has a nice description on this: Allocating Memory for User Contexts (UNIX) - SAP Memory Management (BC-CST-MM) - SAP Library
- If all the standard tricks (block processing, freeing up internal tables...) don't work it really might be an option to push down more logic to the database. This really depends on the kind of calculations you have to do. Yes, HANA is based on this idea. But to some extend you can follow this path with other databases, too. And it even might consume less time on the database server if you don't need to transfer several millions of rows. But once again, this really depends on the calculations you need.
Regards,
Jan