Capacity and SKUs in Power BI embedded analytics
When moving to production, Power BI embedded analytics requires a capacity (A, EM, or P SKU) for publishing embedded Power BI content.
Capacity is a dedicated set of resources reserved for exclusive use. It offers dependable, consistent performance for your content.
For publishing, you'll need one Power BI Pro or Premium Per User (PPU) account.
What is embedded analytics?
Power BI embedded analytics includes two solutions:
Power BI Embedded - Azure offering
Embedding Power BI as part of Power BI Premium - Microsoft Office offering
Power BI Embedded
Power BI Embedded is for ISVs and developers who want to embed visuals into their applications.
Applications using Power BI Embedded allow users to consume content stored on Power BI Embedded capacity.
Power BI Embedded recently released a new version, called Embedded Gen2. Embedded Gen2 simplifies the management of embedded capacities, and improves the Power BI Embedded experience. For more information, see Power BI Embedded Generation 2.
Power BI Premium
Power BI Premium is geared toward enterprises who want a complete BI solution that provides a single view of its organization, partners, customers, and suppliers.
Power BI Premium is a SaaS product that allows users to consume content through mobile apps, internally developed apps, or at the Power BI portal (Power BI service). This enables Power BI Premium to provide a solution for both internal and external customer facing applications.
Capacity and SKUs
Each capacity offers a selection of SKUs, and each SKU provides different resource tiers for memory and computing power. The type of SKU you require, depends on the type of solution you wish to deploy.
To understand which workloads are supported for each tier, refer to the Configure workloads in a Premium capacity article.
To plan and test your capacity, use these links:
Power BI Embedded SKUs
Power BI Embedded is shipped with an A SKU.
Power BI Premium SKUs
Power BI premium offers two SKUs, P and EM.
Which SKU should I use?
The table below provides a summary of features, the capacity they require, and the specific SKU that is needed for each one.
|(A SKU)||(P and EM SKUs)|
|Embed for your customers(app owns data)||✔||✔|
|Embed for your organization(user owns data)||✖||✔|
|Microsoft 365 apps(formerly known as Office 365 apps)||✖||✔|
|Secure URL embedding(embed from Power BI service)||✖||✔|
- A Power BI Pro or Premium Per User (PPU) license is needed for publishing content to a Power BI app workspace.
- Only the P SKU allows free Power BI users to consume Power BI apps and shared content, in Power BI service.
The table below lists payment and usage considerations per capacity.
|Payment and usage||Power BI Embedded||Power BI Premium||Power BI Premium|
|Commitment||None||Yearly||Monthly or yearly|
|Usage||Azure resources can be:
||Embed in apps, and in Microsoft applications||Embed in apps, and in Power BI service|
SKU memory and computing power
The table below describes the resources and limits of each SKU.
|Capacity SKUs||Total v-cores||Backend v-cores||Frontend v-cores||Max memory per dataset (GB)1, 2, 3||DirectQuery/Live connection (per second)1, 2||Max memory per query (GB)1, 2||Model refresh parallelism2|
1 The Power BI Premium Utilization and Metrics app doesn't currently expose these metrics.
2 These limits only apply to the datasets workload per capacity.
3 The Max memory per dataset (GB) column represents an upper bound for the dataset size. However, some memory must be reserved for operations such as dataset refreshes and queries. The maximum dataset size permitted on a capacity may be smaller than the numbers in this column.
4 SKUs greater than 100 GB aren't available in all regions. To request using these SKUs in regions where they're not available, contact your Microsoft account manager.
Embedded Gen 2 memory enhancements
The amount of memory available on each node size is described in the RAM (GB) column in the SKU memory and computing power table. With Power BI Embedded Generation 2 (also known as Embedded Gen 2), it's set to the memory footprint limit of a single Power BI item (such as a dataset, report or dashboard), and not to the cumulative consumption of memory. For example, in an Embedded Gen2 A4 capacity, a single dataset size is limited to 25 GB, compared to the original Power BI Embedded capacity, where the total memory footprint of all datasets handled at the same time was limited to 25 GB.
Submit and view feedback for