on 01-20-2015 1:24 PM
Hi,
I am doing an simple operation where i am mapping fields form EDIDS-SAP table to a temp table , Please note that i have not used any filters in the operation.
Its taking more time and i have tried increasing array fetch but still no luck ..Is there any other points i need to consider and try ?
Thanks,
Arun.
Hi Dirk & Surya,
My DS version is 4.1 and i dont have an option to use abap data flow so going with the normal data flow.
And i am extracting only 8 columns out of the total ones but still its taking close to 45 mins ..
Any leads pls.
Thanks,
AJ.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
You use CREDAT (and CRETIM) columns in a where-clause.
Suppose you build a daily job, then every day you can load the IDOCs generated the day before:
In DS 4.2 the condition is pushed to the underlying database. That means only yesterday's data will be sent over to DS, which will speed up your process enormously.
Hi,
I would like to know how it works in the back end ? I assume that the moment i pull the EDIDS table in designer it will have all the records on it and based on the where condition i apply it will be filter out the records or when ever i execute the query only the EDIDS will get filled with the records ?
If EDIDS only getting filled when the query is executed , I already have all the IDOC numbers in a table and i can use that IDOC numbers in EDIDS table where clause to pull the relevant fields ? Rather than going with the date option in the global variable !!
Thanks,
AJ.
Message was edited by: Arun Joseph
Hi,
you can use RFC_READ_TABLE to read using certain where clauses. Even BODS use the same functionality with BODS_RFC_READ_TABLE , these two functions are same.
In the former one you can filter the records from sap end where as in latter one DS will pull entire data into DS and then filter it which will be error prone as the volume grows huge.
If the records in the table are in lakhs, then the pull time may go more than 30 minutes where the entire job will fail.so better not to go with entire data pull;
Thanks
Phaneendranadh
Arun,
I know you have already mentioned that ABAP Dataflows are not an option for you but scenarios like you mentioned are exactly where ABAP DFs need to be used. Not only is your development time significantly lower (Because it is a straightforward where clause pushed down to ECC) but also you will achieve performance which you cannot match with regular dataflows.
I haven't used the 'RFC_READ_TABLE' function myself but below is a link to the SAP supplement. Page 95 describes how to invoke a RFC function in general.
http://help.sap.com/businessobject/product_guides/sbods42/en/ds_42_sap_en.pdf
Good luck,
Chaitanya
that's a huge table and i sssume it has 25 columns. did you try using ABAP data flow ? can you see how many records this table has in your SAP system ?
Thanks
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Which DS version? Are you using ABAP or normal data flow?
You can reduce the load volume (and as such increase the speed) by removing columns you won't need from the extract, if any.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
84 | |
25 | |
12 | |
9 | |
6 | |
6 | |
5 | |
5 | |
4 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.