Dear Experts,
We have a situation whereby we are required to redesign our complete cube dimensions due to loading performance issue. We have the 3.x flow and Lately it's been taking 7 hours just to load a couple thousand records from PSA to cube and it's been painful.
Now, in order to redesign the cube we will need to delete the cube data and since we have all the data present in PSA we can load from this as starting point. However, we have about 6-7 years of data stored in psa which will need to be loaded to cube.
In this case, since we do not have the convenience of loading via dtp unfortunately I would like to know the fastest or rather the best approach to do so considering the huge amount of data n no.of requests and no. Of datasources as well which provide data to this cube.
Could you please let me know how we can go about this.
Appreciate any help I can get. Thanks much.
Jem