The ultimate Microsoft Fabric, Power BI, Azure AI, and SQL learning event: Join us in Stockholm, September 24-27, 2024.
Save €200 with code MSCUST on top of early bird pricing!
Find everything you need to get certified on Fabric—skills challenges, live sessions, exam prep, role guidance, and more. Get started
Hello Community,
I have been experiencing really slow refresh in Power BI desktop with all my sources coming exclusively Dataflows in Power BI Service. Taking more than 3 hours in one table only.
Below things thay I have already tryed, with same results.
What I can't understand is that same semantic model published on Power BI server takes less than 10 minutes to fully refresh, so the problem seems to appear only when I work in Power BI Desktop.
I appreciate your suggestions ans ideas to solve this issue.
Query Folding: Ensure that your queries support query folding. Query folding means that Power Query can translate transformations into source queries, reducing the amount of data processed locally. This can significantly speed up data refreshes. Use tools like step folding indicators in Power Query Online to verify .
Incremental Refresh: Implement incremental refresh in your dataflows. This approach refreshes only the data that has changed rather than the entire dataset, reducing refresh time. This method is particularly useful for large datasets.
Dataflow Optimization: Optimize your dataflows by separating the ETL (Extract, Transform, Load) process into different stages. Use staging dataflows to handle initial data extraction and transformation dataflows for subsequent transformations. This separation can improve performance by allowing parallel processing and better resource management.
Enhanced Compute Engine: If you are using Power BI Premium, ensure that the enhanced compute engine is enabled. This engine accelerates query processing and refresh operations by leveraging SQL Cache, which can significantly speed up performance.
Performance Best Practices: Follow best practices for developing dataflows, such as using linked and computed entities to reuse data and reduce redundancy, and performing complex transformations only when necessary. Avoid excessive use of expansive operations like joins and merges on large datasets.
Capacity Management: If you're experiencing slow refresh times due to resource constraints, consider scaling up your Power BI Premium capacity. Adjust the size of your capacity and manage refresh schedules to avoid peak times, which can help distribute the computational load more evenly.
Network Latency: Minimize network latency by ensuring your data sources, dataflows, and Power BI services are located in the same region. This can reduce the time required for data to travel between different components of your infrastructure.
By implementing these strategies, you can optimize your Power BI Desktop refresh times and improve overall performance. If the problem persists, you might also consider reaching out to Power BI support for further assistance.
Hi @AndresPesca ,
Base on your description, it seems like the report refresh is very slow in Power BI Desktop. However, it will be faster when published it to Service. And you've already taken substantial steps to troubleshoot the issue, which is great. In order to make further troubleshooting, could you please provide the following info?
Best Regards
Hello,
Thanks for your repply and for your help. Find my answer below in blue.
Hi @AndresPesca ,
The version of your Power BI Desktop is the latest one. As the description in your original post, it seems like the dataflow only include one table. Could you please provide the following info for that dataflow?
Understand and optimize dataflows refresh - Power BI | Microsoft Learn
Best Regards
User | Count |
---|---|
121 | |
108 | |
101 | |
47 | |
38 |
User | Count |
---|---|
192 | |
82 | |
80 | |
72 | |
69 |