Forum Discussion
Power BI Users and Fabric CU Planning
I have a site that uses SQL Reporting Services. There are approximately 45 users, and as part of a migration to Dynamics BC they are considering remaking their reporting service site in Power BI. We were previously looking and debating whether they would need the Power BI Pro or Premium licenses as there is some need for paginated reports (this is still to be defined).
That said, I understand power bi licensing is now shifting to fabric licensing and it is based on CU's or capacity units. Any recommendations or thoughts as to how we may estimate the necessary CU's to support this site? Can the CU's be considered comparable to SQL core licensing? Beside's the CU's are there other sku's we should consider? Right now this is just rough estimating so we can begin setting some expectations. Look forward to a response.
JG
Hi LongtimeGPandSL - When transitioning to Power BI and considering Fabric licensing, it’s crucial to assess several factors regarding Capacity Units (CUs) and their potential usage.
Capacity Units are a measure of the compute and storage resources available in a Power BI Fabric capacity. Each workload (Power BI reports, datasets, paginated reports, dataflows, etc.) consumes CUs based on its complexity and the size of operations.
CUs are conceptually similar to SQL core licensing in that they represent the computational power allocated to your workloads. However, the consumption model for Fabric is more dynamic, as it adjusts based on the type and intensity of usage across multiple services (Power BI, Data Factory, Synapse, etc.).
The number of users and expected concurrent usage. For 45 users, a starting point could be a smaller Fabric Premium SKU (e.g., P1 or P2, equating to 32-64 CU).
Paginated reports: Each report run could consume 1-2 CU per execution depending on complexity.
Additional workloads: Data refresh schedules, AI/ML tasks, and live connections.Storage Needs: Premium SKUs provide more storage compared to Pro licenses (10 GB/user for Pro vs. pooled storage in Premium).
Fabric Workloads: If planning to use Synapse, Data Factory, or Lakehouse, ensure the capacity aligns with your broader architecture.
1 Reply
- rajendraongole1
Super User
Hi LongtimeGPandSL - When transitioning to Power BI and considering Fabric licensing, it’s crucial to assess several factors regarding Capacity Units (CUs) and their potential usage.
Capacity Units are a measure of the compute and storage resources available in a Power BI Fabric capacity. Each workload (Power BI reports, datasets, paginated reports, dataflows, etc.) consumes CUs based on its complexity and the size of operations.
CUs are conceptually similar to SQL core licensing in that they represent the computational power allocated to your workloads. However, the consumption model for Fabric is more dynamic, as it adjusts based on the type and intensity of usage across multiple services (Power BI, Data Factory, Synapse, etc.).
The number of users and expected concurrent usage. For 45 users, a starting point could be a smaller Fabric Premium SKU (e.g., P1 or P2, equating to 32-64 CU).
Paginated reports: Each report run could consume 1-2 CU per execution depending on complexity.
Additional workloads: Data refresh schedules, AI/ML tasks, and live connections.Storage Needs: Premium SKUs provide more storage compared to Pro licenses (10 GB/user for Pro vs. pooled storage in Premium).
Fabric Workloads: If planning to use Synapse, Data Factory, or Lakehouse, ensure the capacity aligns with your broader architecture.