New Power BI Aggregations Feature
Using aggregations in Power BI enables interactive analysis over big data in ways that previously were not possible. Aggregations can dramatically reduce the cost of unlocking large datasets for decision making. The advantages of this are:
- Query performance over large datasets - as users interact with visuals on Power BI reports, DAX queries are submitted to the dataset. Boost query speeds by caching data at the aggregated level, using a fraction of the resources required at the detail level. Unlock big data in a way that wouldn't otherwise be possible.
- Data refresh optimization - reduce cache sizes and refresh times by caching data at the aggregated level. Speed up the time to make data available for users.
- Achieve balanced architectures - allow the Power BI in-memory cache to handle aggregated queries, which it does effectively. Limit queries sent to the data source in DirectQuery mode, helping stay within concurrency limits. Queries that do get through tend to be filtered, transactional-level queries, which data warehouses and big-data systems normally handle well.
The aggregations feature is in Preview, and must be enabled in Power BI Desktop. To enable aggregations, select File > Options and Settings > Options > Preview Features, then select the composite models and the Manage aggregations checkboxes.