Forum Discussion
Power bi large semantic model refresh
Hello everyone,
I hope you're all doing well. I'm writing to seek assistance with an issue I've encountered regarding the Semantic Model First Refresh in Power BI.
I work at a call center company, so we deal with data from millions of calls per day. I have a project to monitor calls, so my semantic model already has more than 60 million lines, with a history only from December 2024 to the present.
It is configured as a large semantic model and I have configured incremental updates. I can publish the application without any problems, but when I try to do the first refresh through an on-premises data gateway, the process crashes while trying to do the first load with the historical data, and I have not yet been able to find a solution.
To have the data updated, I am having to update the entire model in PBI Desktop and then republish the updated project, but as the database grows, it will become unfeasible to maintain this process.
Any suggestions?
You can bootstrap the initial refresh using a tool like SSMS. The Guy in a Cube folks did a video on this: https://www.youtube.com/watch?v=5AWt6ijJG94
Hi MarquesLJ
I would recommend upgrading the CPU for the gateway.
Another option is to make sure that you can reduce any steps in Power Query which could be using a lot of CPU and memory such as sorting or Merges.
7 Replies
- blopez11Super User
You can bootstrap the initial refresh using a tool like SSMS. The Guy in a Cube folks did a video on this: https://www.youtube.com/watch?v=5AWt6ijJG94
- MarquesLJRegular Visitor
Hey blopez11 I'll give it a try. But I've already tried XMLA, but even tho my license is Premiu Per User, when I try to enable XMLA it says I need to buy the premium capacity
It's im Portugues but it reads: Need more control over performance and scale?
Purchase Premium capacity, an add-on to Power BI Pro, to scale your BI solutions and easily distribute content to everyone in your organization.
- v-kathullacCommunity Support
Hi MarquesLJ ,
As we haven’t heard back from you, we wanted to kindly follow up to check if the solution provided for the issue worked? or Let us know if you need any further assistance?
If our response addressed, please mark it as Accept as solution and click Yes if you found it helpful.
Regards,Chaithanya.