on 09-08-2008 6:41 AM
Hi Guru's,
Can somebody please help me, in understanding the query design when we are using it as a source to apd. I have a scenario where i do some transformation using abap and then send the output to a pc file. However, i have huge data in the underlined multiprovider of a query( say 20 million) and the query is getting timed out.
The only way is to have query variables to be able to restrict the data. When I do that, the APD is not supporting. Did somebody came across these scenario's.
Any ideas will be highly appreciated.
Thanks.
Hi,
can you please tell me if you are using BI oder BW ? In BI you can use variables in filter, in BW you can't.
If you have problems in an APD to get all the required data, check, if you have limited your load to only required fields. Always keep in mind that each step of an APD is done in memory. So if your system cannot provide enough memory, it starts swapping and become slow. To avoid this, try working with minimal fields.
If at all possible do not use a Query as Input. Use the Cube instead and apply filters in the next step. It seems that optimization then only collects the required data.
ABAP Steps are not of much use as a filter, because for them the whole content of the data source is read. So all is in memory, and only then the ABAP coding including the filter will apply.
Hope this helps.
Kind regards,
Jürgen
Edited by: Jürgen Kirsch on Sep 8, 2008 9:46 AM
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
87 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.