We are excited to unlock additional asynchronous refresh capabilities for Power BI Premium, Power BI Premium per User, and Power BI Embedded! As Power BI becomes a superset of Azure Analysis Services, the asynchronous refresh REST API for Power BI brings built-in features for reliability, such as auto retries and batched commits. It is a significant improvement to the existing Power BI refresh capabilities, providing enterprise grade refresh capability for your largest datasets!
A critical requirement for most enterprise BI environments is an up-to-date data model. Due to long processing times or network interruptions, it can be a challenge to refresh your largest datasets. Client tools and processes such as SQL Server Management Studio (SSMS), PowerShell cmdlets, Tabular Object Model (TOM), or Tabular Model Scripting Language (TMSL) provide fine-grain dataset management capabilities inherited from Azure Analysis Services for semantic models. However, performing long-running refresh operations using these tools requires maintaining an ongoing HTTP connection. If either the client or network is interrupted, the processing operation is cancelled. Though there has been significant investment to ensure the stability of these methods, for your large datasets, it is more reliable to avoid long-running HTTP connections from client applications. Asynchronous processing eliminates the need for a synchronous client connection increasing processing reliability and enabling the scalability of your big data models.
Please visit our documentation page for details on how to use the REST API for Power BI. It covers how to perform asynchronous refreshes, check their status, and cancel them if necessary. Additionally, the C# RestApiSample code sample is provided.