on 02-28-2008 8:48 PM
Hi
Whats the general method everyone uses to load the PA from a Infocube ? Let me explain more..
We get the data everyday into a cube, will the same cube be used to load it again and again or move the contents of this cube to another cube, and only the delta data is loaded into PA (we use ADD DATA here). If we use the same cube, we load the same data everyday, which takes more time in the job.
Can anyone pls explain the best solution ? From a external BW system infocube, the data is moved to APO BW infocube and loading the PA. and next day how the delta is taken care ?
thx
venkat
Hi Srinivas,
Your process looks bit complex. So everyday you do flatfile comparison for thousands of records ? You do it daily or weekly load ? Who will do this comparison ? with previous data in Infocube ?
Now I realise there are various methods in practice. Anyway thanks srinivas for your inputs. Assigning points..
thx
venkat
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Srinivas,
thx for the reply. If you blank it out, then are you loading the same data everyday with little delta adding up, everyday ?
or if you are using delta in cube, then are you using ADD DATA in TSCUBE ?
One of these should happen.
thx
venkat
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
We do replace the data, hence it is UPDATE mode of delta data. So we use Flatfile comparison tools to check with the last load and then only create a delta file with new CVCs or those CVCs for which values have changed. These are then loaded through infocube in replace mode. We found it was more efficient to use flat file text comparison tools than do the comparison in APO or BW.
Typically the load data that we do is always delta. This means that the Infocube is only used a conduit to get data into APO. Usually data such as History does not change in buckets that are well into the past. So the delta update into the Infocube may contain data only in the immediate past and it may also contain new CVC related data. We typically blankout the infocube before loading using the "Delete Entire Content" option under the Data Targets view of Infopackage.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
9 | |
4 | |
3 | |
3 | |
2 | |
2 | |
1 | |
1 | |
1 | |
1 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.