Forum Discussion
Power BI Scaling Limits
A client of mine is working with SQL Server, Analysis Services Tabular Model, and Power BI (currently Pro and Desktop). There is an emerging use case that would require scaling the data model (which currently utilizes roughly 20 tables) to have one of the tables scale to 200 million rows, one to scale to 60 million rows, a third to scale to 120 million rows, and a fourth to scale to about 10 million rows. The remaining 12-16 tables will be relatively small (hundreds to thousands of rows). The 200 million row table might have as many as 50-70 columns but we are looking to try to shrink the columns to 25-50 if we can. The other tables will typically have 25-50 columns each.
We plan to use SSAS Tabular or Azure Analysis Services Tabular and to push column definition up to the database layer.
There is concern on the team that SQL Server, Analysis Services, and Power BI cannot scale to this level for practical use.
Are there opinions and or material from Microsoft on this topic I can provide my client?
Are there technical and/or practical scaling limits for a data
2 Replies
- v-xuding-msft
Community Support
Hi keobrie ,
There are blogs about comparison of different mode that may help you a little bit.
Power BI Tips: Import vs Direct Query
Power BI – Live connection vs. Import, comparison and limitations
Best Regards,
Xue Ding
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
- keobrie
Helper I
Adding this response because the blogs did not answer my question. Hopefully Microsoft representatives will see this as I understand scalability and best practices with very large data sets are issues for many.
To the extent Microsoft can provide a document (or someone can point me to a Microsoft document) that covers this information, I would be grateful.