Which Fabric capacity SKU should be recommended to minimize costs for a deployment that includes 2,000 Power BI consumers without an individual Power BI Pro license, 32 Power BI modelers with a Pro license, and 16 data scientists?

• A. F32
• B. F64
• C. F128
• D. F16

Answer: (B)

Sizing Fabric capacity SKUs centers on providing enough compute and memory to serve a large pool of users who don’t hold Pro licenses, while still supporting authors and data science workloads at a reasonable cost. In this scenario, 2,000 Power BI consumers without Pro licenses can view content only if it’s deployed on Premium-capacity, so the capacity tier must handle a high level of concurrent access. At the same time, there are 32 modelers with Pro licenses who publish and refresh content and 16 data scientists who may run notebook or analytics workloads, all contributing to the overall demand on the capacity.

The F64 tier offers a balanced level of compute and memory that can accommodate thousands of concurrent viewers plus the additional authoring and data science workloads without jumping to the most expensive tier. The smallest tier would likely struggle with 2,000 simultaneous viewers; the next-up tier could still be insufficient under concurrent authoring and data-processing pressure and would cost more than needed; the largest tier would meet the demand but at unnecessary expense. Therefore, the F64 capacity SKU provides the best mix of performance and cost for this deployment.