Premium Workspace Impact on Creating/Updating Dashboards

[heg37]

Hi, 

I'm not really sure how to ask the question/address the issue I'm having but will try to explain. I have a dashboard that is published to a Premium Workspace (that is shared across my team and includes many other projects saved by others). When I am trying to make updates and changes the data that feeds this dashboard, however, on PowerBI Desktop in the file itself, I am frequently running into memory problems and timeout when I try to do things like create a conditional new column, etc. My data is very large and will only ever exponentially grow, and I need to be able to handle the data in PowerBI Desktop. When I look up the memory error I receive, some forums say to upgrade to Premium. Are there different types of Premium that refer to the individual user's account versus a workspace? How does Premium impact my ability to handle large data in creating new measures with conditional columns, updating visuals etc. without running out of memory? I do 99% of my processing on the data prior to uploading into PBI, but there are some incremental things that come up in visualization that prompt new measures or columns and cause this problem. Help please?

[v-junyant-msft · Accepted solution]

Hi @heg37 ,

@Idrissshatila Thanks for your concern about this case!

Firstly, it's important to distinguish between Power BI Premium at the workspace level and Power BI Premium Per User (PPU). Power BI Premium provides dedicated capacity to your organization, meaning that resources are not shared with other customers. This can significantly improve performance for datasets, reports, and dashboards hosted in Premium workspaces. On the other hand, Power BI Premium Per User (PPU) is a per-user license that offers access to Premium features on a per-user basis, without the need for dedicated capacity.

And here are a few considerations and steps you can take to address the memory issues and timeouts:
1. Using calculated columns judiciously, as they can significantly increase the memory footprint of your model.
2. Leveraging measures over calculated columns when possible, as measures are calculated at query time and do not consume memory in the same way.
3. Reviewing and optimizing your data model for unnecessary columns and tables. Every bit of optimization can help reduce the overall memory footprint.

If you choose a Premium workspace, you have access to features that can help manage large datasets more effectively.
1. Aggregations can reduce the amount of data that needs to be processed and stored in memory by summarizing detailed data into a smaller set.
Automatic aggregations overview - Power BI | Microsoft Learn
2. Incremental Refresh: This feature allows you to refresh only the data that has changed, rather than the entire dataset, which can help manage large datasets more effectively.
Incremental refresh for semantic models and real-time data in Power BI - Power BI | Microsoft Learn

And about Premium Per User (PPU) for Development: If you're not already using PPU, consider utilizing it for development work. PPU gives an individual user access to Premium capabilities, which can be beneficial for development and testing before publishing to a Premium workspace.

Best Regards,
Dino Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Idrissshatila]

HEllo @heg37 ,

this depends on the way you're storing the data, make sure to remove data that you don't need in your analysis.

also build your model as a starschema data model https://learn.microsoft.com/en-us/power-bi/guidance/star-schema

if data can be grouped by, the do grouping to minimize the data.

check this for power bi optimization techniques https://www.linkedin.com/pulse/power-bi-optimization-framework-ahmad-chamy-nensf/?trackingId=SlTNgQF...