Resource Governance: This operation was canceled because there wasn't enough memory to finish

[Murtaza_Ghafoor]

I have f8 capacity & i am getting this error, also this option is enabled
Large semantic model storage formatFor most Premium capacities, using large semantic model storage format can improve performance.

Don.t know what else to do, kindly suugest some workaround.

 
Data source error: Resource Governance: This operation was canceled because there wasn't enough memory to finish running it. Either reduce the memory footprint of your dataset by doing things such as limiting the amount of imported data, or if using Power BI Premium, increase the memory of the Premium capacity where this dataset is hosted. More details: consumed memory 2819 MB, memory limit 2747 MB, database size before command execution 324 MB.

[v-sathmakuri · Accepted solution]

Hi @Murtaza_Ghafoor ,

Thank you for reaching out to Microsoft Fabric Community.

The error is caused by exceeding the memory limits enforced by the Fabric (F8) capacity. As per Microsoft documentation, these limits apply not only to dataset size but also to memory consumed during query and refresh execution, which can be significantly higher due to intermediate operations.

Enabling Large Semantic Model storage format does not increase the available execution memory; it only affects storage capabilities.

To resolve this,

• Reduce the dataset memory footprint (remove unused columns, lower the cardinality, optimize model design)
• Implement incremental refresh to avoid full dataset processing
• Optimize DAX queries to reduce intermediate memory usage
• Follow star schema modeling practices

If the workload still exceeds the limits after optimization, the recommended approach is to scale the capacity to a higher SKU (e.g., F16) as F8 may not provide sufficient memory for this workload.

Thanks!!

[v-sathmakuri · Accepted solution]

Hi @Murtaza_Ghafoor ,

Thank you for reaching out to Microsoft Fabric Community.

The error is caused by exceeding the memory limits enforced by the Fabric (F8) capacity. As per Microsoft documentation, these limits apply not only to dataset size but also to memory consumed during query and refresh execution, which can be significantly higher due to intermediate operations.

Enabling Large Semantic Model storage format does not increase the available execution memory; it only affects storage capabilities.

To resolve this,

• Reduce the dataset memory footprint (remove unused columns, lower the cardinality, optimize model design)
• Implement incremental refresh to avoid full dataset processing
• Optimize DAX queries to reduce intermediate memory usage
• Follow star schema modeling practices

If the workload still exceeds the limits after optimization, the recommended approach is to scale the capacity to a higher SKU (e.g., F16) as F8 may not provide sufficient memory for this workload.

Thanks!!

[Murtaza_Ghafoor · Accepted solution]

Yes i already fixed this issue yesterday by reducing the number of queries during refresh time.Thanks for your input.

[v-sathmakuri]

Hi @Murtaza_Ghafoor ,

Thank you for confirming that the issue is resolved.