Why can only a Compute instance of Azure Machine Learning be assigned to one user

Hi zmsoft

Welcome to Microsoft Q&A Forum, thank you for posting your query here!

A Compute instance in Azure Machine Learning is designed to be dedicated to a single user who has root access. This single-user assignment ensures that all actions performed on the compute instance use that user's identity for Azure RBAC (Role-Based Access Control) and attribution of experiment jobs. This design is primarily for security and management reasons, allowing for a controlled environment where each user can manage their own resources without interference from others.

Sharing a compute instance between multiple users will disrupt the previous configuration and setting applied by other user and can also create issues like OOM, thread locks etc. Instead, you can opt for a  compute clusters to share amongst users which scales automatically based on scale up time and loads.

kindly refer the below Document: https://zcusa.951200.xyz/en-us/azure/machine-learning/how-to-manage-compute-instance?view=azureml-api-2&form=MG0AV3&tabs=python#manage

Hope this helps. Do let us know if you any further queries.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

Thank You.