SkyPilot: Run AI and batch jobs on any infra (Kubernetes or 12+ clouds). Get unified execution, cost savings, and high GPU availability via a simple interface.
I have been messing around with skypilot the past few days (really awesome, work, I love it), and I noticed that if I delete the ~/.sky directory that sky is not able to track the services anymore (and I have to manually find and delete the running instances). Is there a way to save this information in a recoverable way? Ideally a way that can be shared with others? I think being able to save the database to git would also work. Is it possible to configure a directory / path to save essential databases such that I can track it with git in my project? Another idea would be that can't it recover the services based on the instances that are running? e.g. the instances running ec2?
I have been messing around with skypilot the past few days (really awesome, work, I love it), and I noticed that if I delete the ~/.sky directory that sky is not able to track the services anymore (and I have to manually find and delete the running instances). Is there a way to save this information in a recoverable way? Ideally a way that can be shared with others? I think being able to save the database to git would also work. Is it possible to configure a directory / path to save essential databases such that I can track it with git in my project? Another idea would be that can't it recover the services based on the instances that are running? e.g. the instances running ec2?