> Compute Engine may shut them down after providing you a 30-second warning, and you can use them for a maximum of 24 hours.
That's a good start...looks like the model is "save your work often". I'm sure you could run a few instances for a while and model the distribution of time granted.
Right, my concern is more the event of them axing work 32 seconds in or something, which could be a problem for some data analytics tools I've seen, that need a good 70 seconds or so to reach a steady state.
Quoting their docs: "Compute Engine doesn't bill for instances preempted in the first 10 minutes," so that scenario is harmless in most application designs where preemptible VMs would make sense anyway.
Disclosure: I used to work at Google on GCE, but not directly on preemptible VMs or this billing policy. I don't currently work or speak for Google.
That's a good start...looks like the model is "save your work often". I'm sure you could run a few instances for a while and model the distribution of time granted.