I would like to process several cubes within one job - let's use example: cube A, cube B, cube C, all located on same SQL Server, similar cubes for different customers. Is there a way how to do it easily, without too much manual effort? E.g.: definition of several cube names in XMLA script or using lookup in ETL job (SSIS), which would match cube name with data in the table? Or some other way. I just would like to avoid situation where I need to duplicate steps in ETL job (for each cube processing separately). Thanks in advance!
Asked
Active
Viewed 497 times
1 Answers
2
If you have multiple Cubes under the same Database, you can process at the Database level itself. In this case, it will process all the Cubes.
For e.g., in the AdventureWorks sample, there are 2 Cubes:
The xmla approach provides the simplest way to process. If you run the following, it would process the Analytics database.
<Batch>
<Parallel>
<Process>
<Object>
<DatabaseID>AW2014</DatabaseID>
</Object>
<Type>ProcessFull</Type>
<WriteBackTableCreation>UseExisting</WriteBackTableCreation>
</Process>
</Parallel>
</Batch>
But, if your Cubes are under different databases, you can generate the above xmla for different databases and run them one by one or in parallel (depending on the size of data / server capacity).

Subbu
- 2,130
- 1
- 19
- 28