Power BI dataflows: building your own refresh orchestration
summary
Shows a working orchestration that runs Power BI dataflows in ordered blocks by calling the Power BI REST APIs from Azure Functions and marshalling execution with Power Automate using JSON metadata and polling for status. Built components: two function endpoints (start + status), a JSON list-of-blocks, retry logic and delays; demo runs six flows correctly. Notes performance change when using Enhanced Compute Engine where a sample run drops from 202s to 48s. Code is on GitHub.
AI-written summary · about
chapters (11)
- 0:00Power BI dataflows intro
- 1:31Live demo: refresh orchestration
- 5:32What are dataflows and limits
- 8:19Orchestrate with Power Automate & Azure Functions
- 12:23Azure Functions code review
- 21:46Power Automate refresh flow
- 27:26Large-scale orchestration
- 36:24Orchestration in action with status
- 36:44Linked entities and compute engine
- 48:51Fabric pipelines for refresh
- 49:35Q&A
related talks
48mGo With The (data)Flow
50mPower BI Incremental refresh made simple.
40mSelf-Service ETL: The PowerBI Data Flows
50mAdvanced Power BI refresh with ADF/Synapse
52mPower Up with the Power Platform
50mDataOps 101 – A Better Way to Develop and Deliver Data Analytics
50mPower BI + Azure Synapse Analytics: Beginning your journey
51mSupercharge Power BI with Azure Synapse Analytics