Weird logging in transaction SAT performance analysis

[EnnoWulff]

Hey there!

When trying to find performance issues, I wonder that routines are called far too often.

The following example is a screenshot of the call order.

at line 177 the function module FUNCTION_EXISTS is called ONE time.

The next command IMPORT FROM MEMORY FUNCTION_LIB is the first command in this function module but is called 4119 times...?!

Am I missing something?

Thanks for any hint.

Cheers
~Enno

[BaerbelWinkler]

Hi Enno,

which tab in the results is the screenshot from and how did you execute SAT parameter-wise? I'm just wondering because FM SD_BUFFER_TABLES_REFRESH in our NW 750 system mostly contains CLEAR- and REFRESH statements and no call to e.g. FUNCTION_EXISTS.

Cheers

Baerbel

[EnnoWulff]

Hey 8b889d0e8e6f4ed39f6c58e35664518f Thanks for your input.

The call hierarchy cannot be seen in this list.

SD_BUFFER_TABLES_REFRESH is called.

Afterwards FUNCTION_EXISTS is called. the entries below are called by FUNCTION_EXISTS.

[BaerbelWinkler]

As far as I know you only really know the sequence of events, if you run SAT with Aggregation set to "None", otherwise the sort order isn't (necessarily) in the sequence of routines called:

With "Per Call Position" you get the number of overall calls for a routine and as one routine can be called from X number of places it can't also show the call sequence.

[EnnoWulff]

hmm. that might be a solution 8b889d0e8e6f4ed39f6c58e35664518f

I am not really sure about it but I will check.

Thanks for that hint!

[EnnoWulff]

8b889d0e8e6f4ed39f6c58e35664518f that was the right hint, thanks!

[BaerbelWinkler]

enno.wulff

Glad I could help! In order to potentially help others as well, I took the liberty to convert my comment to an answer - unfortunately, that didn't move your replies with it.