Premium Version - Power BI Memory Error in 2nd time full refresh. But 1st refresh is a success

[PallaviR]

Dear Team

We have power bi premium version. I am manually publishing the pbix file to the premium workspace. Then refreshing the dataset. It successfully refreshes the dataset giving total size of the data to 3.9 GB.

Then when immediately when I do the second full refresh on the same dataset, it gives me memory error. there is no change in data. How it is giving this error? I am unable to understand when 1st time full refresh is successful, why immediate 2nd full refresh is failing. Has anyone faced this issue, Please help.

Resource Governing: This operation was canceled because there wasn’t enough memory to finish running it. Either increase the memory of the Premium capacity where this dataset is hosted or reduce the memory footprint of your dataset by doing things like limiting the amount of imported data. More details: consumed memory 12065 MB, memory limit 12053 MB, database size before command execution 13546 MB. Learn more, see https://go.microsoft.com/fwlink/?linkid=2159753.

Thanks,

Pallavi

[v-zhangti · Accepted solution]

Hi, @PallaviR 

According to your description, I think that your workspace contains datasets that are too large, which leads to memory allocation failure. You should keep in mind that datasets stored in pbix files are highly compressed.

Therefore, I suggest you take measures against the dataset which causes this problem. You can try these measures:

1. Go to the power query editor in Power BI desktop, delete the useless columns. Then publish it to service to replace the original one.
2. Stagger the refresh time, try to refresh large datasets in free time.

For more information:

https://windowsreport.com/memory-error-allocation-failure/

Solved: Memory error: Memory Allocation failure. - Microsoft Power BI Community

Solved: Memory error: Memory Allocation failure . Try simp... - Microsoft Power BI Community

Best Regards,

Community Support Team _Charlotte

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

[lbendlin]

you are doing expensive Power Query transforms that increase the (temporary) size of your dataset threefold, thus exceeding the 10GB limit. Consider using incremental refresh, or elastic capacity ($$$).

What's your capacity SKU, P1 ?