Add VGPU documentation #44
Conversation
|
|
||
| Host OS configuration is done by using roles in the `stackhpc.linux <https://github.com/stackhpc/ansible-collection-linux>`_ ansible collection. | ||
|
|
||
| Add the following to your ansible ``requirements.yml``: |
There was a problem hiding this comment.
Could these changes go into s-k-c?
There was a problem hiding this comment.
Potentially. I was thinking this first iteration would just contain the documentation to use the role and then we eventually land the playbook in kayobe. We'd then update these docs to use to the kayobe support. What do you reckon?
There was a problem hiding this comment.
For me it depends on when we expect to get it into Kayobe. If we don't think we'll have it in time to backport to the (downstream) Antelope release, then I'd suggest we commit the custom playbook. It can always be removed in favour of the Kayobe support in future, and keeps everyone in sync.
FWIW, I didn't put the Kayobe VGPU support in the infra roadmap, thinking we'd probably end up moving on after getting to this point. That said, if you think it's a reasonably small piece of work then it might be worth doing in time to backport to antelope downstream.
Co-authored-by: Mark Goddard <[email protected]>
Co-authored-by: Mark Goddard <[email protected]>
|
|
||
| Host OS configuration is done by using roles in the `stackhpc.linux <https://github.com/stackhpc/ansible-collection-linux>`_ ansible collection. | ||
|
|
||
| Add the following to your ansible ``requirements.yml``: |
There was a problem hiding this comment.
Potentially. I was thinking this first iteration would just contain the documentation to use the role and then we eventually land the playbook in kayobe. We'd then update these docs to use to the kayobe support. What do you reckon?
Co-authored-by: Mark Goddard <[email protected]>
Adds documentation for NVIDIA vGPUs.