nogox
Per-second billing doesn't add much value to EC2.
To make the most of per-second billing, the compute unit should be deployed within seconds, e.g. immutable. prebaked container. You launch containers on demand, and pay by seconds.
ZEIT or Hyper – a new Heroku emerging
Ask HN: Does per-second billing make sense for GPU Instance?
Assume that you can provision a GPU instance in EC2 in 10 seconds, and they charge you by seconds, not hours.Does this make sense to your AI/ML project? If the job takes hours or days to finish?
Running Windows Container on Linux Host
Great deck explaining the diff between Docker,CoreOS,Opendeis,Hyper,Joyent,ECS [pdf]
Serverless is only cost effective in low traffic cases. If you get a stable or high volume of request, a traditional server is way cheaper.
They can boot a new VM with Docker images in 200ms, which is very close to LXC, and perfectly isolated by hypervisor.
The problem of "Virtual Machine" is not "Virtual"/Virtualization, the problem is the full blown guest OS, aka "Machine".
It is great to hear you guys are working on that. I find Hyper is pretty interesting to run Docker on Mac https://hyper.sh/blog/post/2015/07/30/running-containers-fro....
I think this is the way Container-as-a-Service to go, instead of the current form of AWS ECS.