With exponential growth in data volumes and complexity, Power BI customers demand enterprise BI solutions that scale to petabytes, are secure, easy to manage, and accessible to all users across their organizations. Such solutions require extensive API surface area for enterprise management scenarios. For example, you can use the Fabric item CRUD APIs to list semantic models, create and update them, and copy them to other workspaces. You can also use the Power BI REST APIs to perform advanced refresh operations and execute DAX queries, among other things. For semantic model metadata operations and refreshes, you can also leverage XMLA endpoints and the Tabular Object Model (TOM) through Analysis Services client libraries.
We are releasing a new Fabric REST API for configuring connection bindings for semantic models. A connection binding defines what data connection a semantic model will use to connect to an underlying data source. In scenarios where data sources change—such as moving from development to production environments, migrating infrastructure, or switching connection details—being able to programmatically update these bindings is crucial. The new Bind Semantic Model Connection REST API marks a significant improvement for BI pros and developers to automate these tasks.
Bind Connection versus Bind to Gateway
The new Bind Semantic Model Connection REST API supersedes the existing Datasets – Bind to Gateway in Group REST API in Power BI. It provides support for all connectivity types, including gateway connections, cloud connections, and default connection settings. It is also possible to unbind a semantic model thus leaving the semantic model in a disconnected state. Moreover, it is fully aligned with the List Connections and List Item Connections REST APIs so that you can work seamlessly across multiple APIs to generate a connections request.
How to configure a connection binding
Here’s a quick rundown of the typical steps to configure a connection binding:
| Task | Comments | |
| Step 1 | List the existing connections for a semantic model | Use the List Item Connections REST API. It returns one or multiple item connections, which correspond to the data sources the semantic model uses. |
| Step 2 | List all connections you can access | Use the List Connections REST API. It returns the connection details for all the connections you have access to. |
| Step 3 | Pick the data connection you want to use for the semantic model binding | Find the connection in the list connection response of Step 2 that matches the connection details of the item connection of Step 1. Take note of the connection id because you most likely will need it in subsequent steps. If there is no matching connection, you can create one by using the Create Connection REST API. |
| Step 4 | Generate the payload for the Bind Semantic Model Connection request | Create a JSON document with the connection binding details, which typically include the connectivityType, the connectionDetails, and the object id of the connection you want to bind. Note that the connection id is not applicable if you want to use default connection settings (“connectivityType”: “Automatic”) or if you want to unbind a data source (“connectivityType”: “None”). |
| Step 5 | Submit the JSON payload in a POST request to the Bind Semantic Model Connection REST API | You must submit a request for each data source individually. To bind multiple data source references of a semantic model, you must submit multiple bind connection requests. |
We’d love to hear from you
The new Bind Semantic Model Connection REST API transforms the connection binding process through seamless automation. Visit the official Fabric REST API documentation as a detailed API reference and use the REST API Try It feature for an intuitive and interactive quick start. We hope that this new REST API for configuring connection bindings lets you streamline your semantic model lifecycle management, deployment automation, and administrative scripts with greater ease and consistency than previously possible.
We value the feedback of our community and are eager to hear your experiences with the new API. Please connect with us on our forums, submit suggestions, or report issues through the official Power BI support channels. Your input will help us shape future enhancements and ensure our APIs meet the diverse needs of our BI pros, partners, and users.