Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'll just answer your driver question: you bring whatever you want. Our VMs just boot off of bytes you have stored in your image (which is in turn just stored on GCS). If those bytes happen to be Debian 8 with some AMD driver that supports the 9300, awesome ;).

We will be working with both vendors to make sure we highlight which drivers work most reliably on our stack though (virtualization + GPUs is way too rare!). We want to work closely with both vendors to do the qualification, so you at least know what is known good.

Disclosure: I work on Google Cloud.



Thanks for the explanation, that's what I guessed -- which has both good & bad aspects. The good is the flexibility, the bad is that:

> If those bytes happen to be Debian 8 with some AMD driver that supports the 9300, awesome ;).

happens to support might alone not be enough given how much difficulties AMD has been having. I just hope that there will be enough customer demand for their SPFlop cruncher GPUs for interested parties -- including Google -- to pitch and contribute to their ROCm stack.


When is Kubernetes support going to be added? Right now it has it only in a very basic form for NVIDIA GPUs.


There is a PR[1] to extend the current alpha GPU implentation. Hopefully it will get merged soon.

[1] - https://github.com/kubernetes/kubernetes/pull/28216


That's still NVIDIA-only, plus you can't even tell a P100 from a K80 for scheduling purposes.


This is the reason node-feature-discovery exists. Kubernetes needs an extensible way to detect and label hardware features: See: https://github.com/kubernetes-incubator/node-feature-discove...


I'd just use a separate NodePool and match your jobs using a selector as appropriate.


So, is gci going to come with the right drivers?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: