2010 Apr 06 8:26 AM
Hi,
I need to update large amount of data into a ztable using INSERT statement.
This task is performance critical.
Which method of table update will have positive impact on the performance -> commit work or call func. in update task.
Thanks for the help.
regards
Karen
the update task is for transactional changes of objects (small number), it creates an overhead which is however necesary to insure data consistency.
If you plan to update really a lot of data, then you should work directly with commit works after around 10.000 records.
You must ensure that nobody works in parallel on your updated inconsistent data.
Siegfried
2010 Apr 06 2:45 PM
the update task is for transactional changes of objects (small number), it creates an overhead which is however necesary to insure data consistency.
If you plan to update really a lot of data, then you should work directly with commit works after around 10.000 records.
You must ensure that nobody works in parallel on your updated inconsistent data.
Siegfried
2010 Apr 07 1:01 PM
If you are updating data based on the outcome of some standard functionality, then use "update task" method. Example is updating the Z table from the Sales order User Exit "MV45AFZZ - USEREXIT_SAVE_DOCUMENT" (Update the ztable only if sales order save is successful) . Otherwise use Direct update and commit work.
Regards
Vinod
| User | Count |
|---|---|
| 6 | |
| 2 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 |