This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
User Data Functions is an item that allows you to create fully managed Python functions to run on Fabric. You can use this item to embed your business logic in your Fabric data architectures across all the different integrations, such as Fabric data sources, Power BI reports, Pipelines, Notebooks, and more.
The use of the Test capability in Fabric User Data Functions operations will use your Fabric capacity beginning October 27th.
In August, we introduced the test capability in Develop mode (preview) to test your User Data Functions directly from your browser. This feature has helped users have faster development iteration cycles, saving them time and increasing the quality of their published functions for other Fabric items to consume.
When you test your functions with this capability, you start a session where you can modify and run your code immediately without having to publish your functions. This session has a timeout of 15 minutes, and it is extended every time you run a new test. The total session time is then reported under the User Data Functions Portal Test operation name in terms of seconds. When this feature was first introduced in preview in August 19th, the User Data Functions Portal Test operation was reported but it did not result in any capacity usage.
With the General Availability release of Fabric User Data Functions, the test capability is now also generally available. Starting on October 27th, the User Data Functions Portal Test operation will start using the capacity reported. There will be no changes to the details or rate in CU/s of the User Data Functions Portal Test operation with this change. This operation will continue to be reported in the Capacity Metrics application as usual.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.