With the rollout of Aggregated Tables in PowerBI, we gained some much-anticipated functionality to improve performance large datasets. The problem I’ve run into is that aggregation is very subjective and can still result in datasets that are still too large for publishing to PowerBI cloud. Additionally, it can take a frustrating amount of time for those aggregate tables to import into a developer’s local instance of PowerBI Desktop. In this article, I’d like to show you how you can leverage PowerBI's Aggregation, functionality, Direct Query and a simple calculation group to partition an aggregated table by Current (import) and Archive (Direct Query) periods. This is one way to work with aggregated tables within the constraints of your licensed capacity and budget. There are definitely caveots to this approach that I will cover along the way.
Read more...
PartitionedAggregateTableExample.pbix