topic Re: Performance issue in Application Development and Automation Discussions

Performance issue

venkatasriram_mygapula — Thu, 17 Jul 2008 07:52:25 GMT

Hi,

when i am executing my program it's working but it's taking 10 min to display the output. in my program i called some archiving function modules to read the data from archiving filis.

its fetcing and display the data from archiving files. but it takes 10 min. how can improve performance on my program.

regards.

sriram.

Re: Performance issue

matt — Thu, 17 Jul 2008 08:13:11 GMT

Do you have a WAIT UP TO 600 SECONDS statement in your program?

Perhaps you need faster disks or processors.

Maybe if you provided some detail, you'd get a sensible answer.

matt

Re: Performance issue

Former Member — Thu, 17 Jul 2008 08:26:57 GMT

hi Shreeram,

This might be due more time taken by Data base.

Go go SE30 tcode ananlyse your program abap time DB timeProcessing time

This might help u

Amit

Re: Performance issue

Former Member — Thu, 17 Jul 2008 16:49:04 GMT

It would have helped if you specified what archiving FMs you are using. Asssuming you are using un-modified SAP delivered FMs, here are some general tips that help:

- apply the widest possible filter to get the least possible records per pass

- break up your data access to several passes. viz. if you are looking to extract 10,000 records, you may want to extract 2500 at a time, in 4 passes. You will need to be creative in doing the break-up

Re: Performance issue

venkatasriram_mygapula — Fri, 18 Jul 2008 02:48:53 GMT

Hi, all

now my program is working with less time compare to previous.

i used some performanc tips like sort internal table use read statement and binary serch. so i could reduce the time.

sekhar as you said in my program some of the function modules related to archiving to read the data taking more time to execute any alternative .

Thanks.

Sriram.

Re: Performance issue

matt — Fri, 18 Jul 2008 07:25:48 GMT

If you are doing lots of READs from a table, and the key is unique, try using a HASHED TABLE rather than using a standard table and doing binary search. It's much quicker.

If the key isn't unique, use a SORTED TABLE. Then you automatically READ in the most efficient way.