PowerBI refresh timeout in Shared capacity

[Asnate]

Hello,

I have issue that after couple hours my Semantic model gets timeout in PowerBI service (Shared Capacity).

Previously it finishet around in 20-30 minutes. Few times in last months it stuck. But after retry it finished successfully. However, in  last two weeks it doesn't load at all in Shared Capacity. Besides more data on daily basis, there aren't any changes in the data model. If I filter 25% of data, then I am able to do data refresh again successfully in like 13 minutes. So I'm not sure what the issue is - why more data causes such problems.

The PowerBI Semantic model size is ~920MB large (pbix file size after the refresh with latest data). So I do not exceed 1GB semantic model limit for PowerBI PRO.

For data storage I use Import mode.

In local PowerBI desktop I can refresh the same dataset successfully.

I can also refresh it Premium per User (Dedicated capacity) in 18-20 minutes.

Any suggestions would be appreciated. Thanks in advance.

[Asnate · Accepted solution]

At first I tried Model optimization, Query optimization etc. Also considered incremental refresh. But all that didn't help since id doesn't load at all then incremental refresh wans't an option.

In the end the problem was multiple calculated colums on a large table. After moving them to Data import steps, it worked like a charm. It seems that PowerBI is hard to manage RAM resource limitations during those calculated column refreshes.

[mcaraballo]

Hello,

Any updates on this?

[v-tsaipranay]

Hi @Asnate ,

Thanks for reaching out to the Microsoft fabric community forum.

@rohit1991's answer correctly addresses solutions to optimize refresh times and performance in Shared Capacity and key solutions include model optimization, incremental refresh, and considering an upgrade to Premium or Premium per User. Additionally, here are other known limitations of Shared Capacity.

• In Shared Capacity, resources are shared among multiple customers, potentially causing slower performance and timeouts during high usage.
• In Shared Capacity, Power BI faces CPU and memory limits, which are lower than those in Premium or PPU.Complex models or high data volumes can lead to longer refresh times, especially during peak periods.
• It has a queue for data refreshes. High user activity can delay or timeout your dataset refreshes due to resource limits, causing slow performance.
• Even though your model is under 1 GB, complex or inefficient queries, large data imports, or complex transformations can strain Shared Capacity resources and cause timeouts.

I am including a resolved thread related to Pro License (Shared Capacity) vs Premium License (Dedicated Capacity). Please refer to it for a better understanding:

https://community.fabric.microsoft.com/t5/Service/Pro-License-Shared-Capacity-vs-Premium-License-Ded...

I hope my suggestions give you good ideas, if you need any further assistance, feel free to reach out.

If this post helps, then please give us Kudos and consider Accept it as a solution to help the other members find it more quickly.

Thank you. 

[v-tsaipranay]

Hello @Asnate ,

I wanted to follow up on our previous suggestions regarding PowerBI refresh timeout in Shared capacity. We would love to hear back from you to ensure we can assist you further.

If my response has addressed your query, please accept it as a solution and give a ‘Kudos’ so other members can easily find it. Please let us know if there’s anything else we can do to help.

Thank you.

[v-tsaipranay]

Hi @Asnate ,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[v-tsaipranay]

Hello @Asnate ,

Could you please confirm if the issue has been resolved? 

If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[mcaraballo]

Hi @v-tsaipranay ,

I have a question regarding this point: "In Shared Capacity, resources are shared among multiple customers, potentially causing slower performance and timeouts during high usage."

Is it possible to know if more limitations were added in Shared Capacity? if the resources in Shared Capacity are less than a month or if for the same resources more customers are competing for?

[v-tsaipranay]

Hi @mcaraballo ,

Thanks for reaching out to the Microsoft fabric community forum and  for actively participating in the community forum and raising a queary in the community forum. Your contributions make a real difference.

As per your question I am inculding you a similer thread which is already resloved so kindly refer to that for better usdersting:

Solved: Shared Capacity Workspace Limits - Microsoft Fabric Community

Thank you.

[mufaza]

I'm having a similar problem. All dataflows were running until 14th of January. Now some of them run successfully occasionally but most of the time they fail. I have already split them up as much as I can. In my case I even have a dataflow that usually did run in 30s but ends up in a timeout after 5 minutes without any explanation.

I'm not using Premium licensing.

[rohit1991]

Hi, @Asnate 
The issue is likely due to Shared Capacity limitations. To resolve it:

1. Optimize Model: Remove unused columns/tables and aggregate data.
2. Enable Incremental Refresh: Refresh only new/changed data.
3. Optimize Queries: Improve source query performance.
4. Upgrade to Premium/PPU: Shared Capacity may lack resources for large models.
5. Schedule Off-Peak Refreshes: Avoid refreshing during busy times.

If issues persist, consider splitting the dataset or upgrading to dedicated capacity.

[mcaraballo]

Hi @rohit1991 !
Hope you are doing well!

I am experiencing the same issue that @Asnate has described above. It seems like people who use Power BI Service on a shared capacity have the same issue since January 11th. So I do not think that just optimizing queries/model would fix this, because it seems to be something that is affecting the shared capacity.

During January 9th and 10th, there was an issue  in Azure that affected some product and services in the East US 2 region. On Saturday morning this issue was solved, but then the schedule refresh of the reports starts failing and keeps happening. could this be related?

[Asnate]

Thank you for your answer. But isn't it still weird that in Premium Capacity data could refresh in 20-30 minutes and the model size is below 1GB. So shouldn't Microsoft provide that model refresh works? Or is there any other known limitations of Shared Capacity besides model size, workspace size and 2h timeout and common pool for PowerBI customers?