Hi All -
Need insight in below cube refresh scenario...We have a cube which is having around 3 Years of data and it is huge. But user requirement is to see all three years data in the same model
This cube refresh is taking long time so we wanted to check for possibilities of going for Incremental refresh using below scenario, need inputs if this can be achieved and if yes how?
Plan is to have 2 MDL files, first MDL file will have time based partition and will refresh full data for first time and create MDC files at Quarter level i.e. (12 MDC files).
As we wanted to refresh only latest quarter information, we will use another MDL file which will generate only the MDC for this quarter, then copy this MDC to the main location where all 12MDC files are created using first MDL.
Will this approach work? If yes, are there any other steps which we need to take to optimize the cube.
Please help on this. Thanks