In the world of containers, I wonder if one could use this feature to interrupt the execution of some computation, save the state somewhere (disk or object storage) and once the workload has been rescheduled on some other node, basically you boot up from your image file, re-establish network connections (if necessary) and just resume computation as if nothing happened.
https://criu.org/ (Checkpoint/Restore In Userspace) is a long-running project (i remember the name over a decade ago) worth checking... Ooh, looks like they made progress in conveniently connecting it to containers.
In the world of containers, I wonder if one could use this feature to interrupt the execution of some computation, save the state somewhere (disk or object storage) and once the workload has been rescheduled on some other node, basically you boot up from your image file, re-establish network connections (if necessary) and just resume computation as if nothing happened.