Savin

@hermesdt Thanks for the feedback! We will add that shortly.

@npow What is the version of Metaflow that you are using? Can you try it with the latest Metaflow version?

@oavdeev We should make the auth optional (but highly suggested) because otherwise, the domain name requirement makes deploying the stack a bit non-trivial.

A quick note on sequencing this PR - we can merge the GCP datastore implementation first, then fast-follow with the changes needed by this PR for Azure and GCP, and...

@kgrvamsi You can use `subprocess.run` to launch a flow from within a python script. #116

@den4uk We are actively thinking about programmatic local execution of Metaflow flows. I will update the thread as we start making further progress with the design memo. Can you elaborate...

@den4uk Metaflow [uses](https://github.com/Netflix/metaflow/search?q=subprocess) `subprocess` behind the scenes and it is very likely that the programmatic execution capability will be a wrapper on top of `subprocess`.

Here is the memo - https://docs.google.com/document/d/1HJW9TH6lHEUojqDTzfgJZJXgjg3xZr43-p8_vjYJM1c/edit#heading=h.thnxk5pwrpsa

It's not just the error in Metaflow runtime - if the container crashes during user code execution, the artifacts are still not going to be persisted (while the flow moves...

@menon92 It does! You can configure spot instances either in [AWS Batch](https://aws.amazon.com/getting-started/hands-on/run-batch-jobs-at-scale-with-ec2-spot/) or in [Kubernetes](https://aws.amazon.com/getting-started/hands-on/amazon-eks-with-spot-instances/).