azuremachine-learningazure-machine-learning-serviceazuremlsdkazure-ml-pipelines

Azure Machine Learning Studio - Data access for different users

am using Azure Machine Learning Studio and have a lot of SQL-Databases and storage accoutns as data sources attached and created many datasets.

Since we've many users (with IAM role Azure Data Scientist) in Azure Machine Learning Studio, we need to distinguish, which user can see use which dataset?

Is there an option for it?

enter image description here

Solution

  • Yes, there's an option to do this.

    Basically the access level for an individual user can't be changed in ML Studio, when the Data asset is already created. But there's an option of different access permissions on the attached storage.

    Therefore the Datastore to create needs identity based data access in order to utilize the users permissions on the storage instead of ML Studios permissions:

    enter image description here enter image description here

    Here's a sample implementation for the above question: enter image description here