How to handle versioning and rollback of a deployed ML model?

Hello @Tú Nguyễn

Thanks for reaching out to us, I think Managed online endpoint in Azure Machine Learning is the closest choice for your requirement.

MLOps is a set of practices and tools that help organizations to manage and deploy machine learning models in a scalable and reliable way. They include cross-functional collaboration, version control and testing, and ensuring the deployment environment is secure and compliant with relevant regulations. By adopting MLOps practices, organizations can improve collaboration between teams, better govern and comply with regulations, and deploy models safely and securely.

The blog with details how Azure Machine Learning can help adopt MLOps practices, with a special focus on model deployment and safe rollout aspect for your reference is here - https://techcommunity.microsoft.com/t5/ai-machine-learning-blog/safely-roll-out-your-machine-learning-models-using-managed/ba-p/3823098

What's covered:

• Azure Machine Learning for model deployment
  • Model versioning and management
    • Deployment techniques for managing reliability
      • Model monitoring and logging
        • Network isolation
• Safe rollout "in action" with Azure Machine Learning
• Deploying a new model to production workspace
• Choose existing endpoint with the old model
• Enable telemetry logging and data exfiltration prevention
• Setting up the Mirrored traffic for the new model
• Finish the deployment and monitor the new model
• Analyze costs and make the final call before initiating the rollout
  • Safely roll out the new model

I hope this information is what you are looking for, please take a look and let us know if you need more information.

Regards,

Yutong

-Please kindly accept the answer if you feel helpful to support the community, thanks a lot.