RHOAIENG-11046 - Add support for AMD GPU image #68
Conversation
support/defaults.go
Outdated
| @@ -8,4 +8,5 @@ const ( | |||
| CodeFlareSDKVersion = "v0.20.2" | |||
| RayVersion = "2.35.0" | |||
| RayImage = "quay.io/modh/ray:2.35.0-py39-cu121" | |||
| RayAMDGpuImage = "quay.io/rhoai/ray:2.35.0-py39-rocm61-torch24-fa26" | |||
There was a problem hiding this comment.
This rocm-ray image quay.io/rhoai/ray:2.35.0-py39-rocm61-torch24-fa26 is torch compatible.
IMO, we should use base Ray AMD/ROCm image"2.35.0-py39-rocm61" built with konflux
quay.io/modh/ray:2.35.0-py39-rocm61
|
Hello @astefanutti can you please help us to find out what is the right image? |
|
@jiripetrlik Yes that's preferable to default to |
jiripetrlik
force-pushed
the
RHOAIENG-11046
branch
from
58428c3 to
7410736
Compare
|
@ChughShilpa @astefanutti Thank you for clarification of the image name! It should be fixed now. |
jiripetrlik
force-pushed
the
RHOAIENG-11046
branch
from
7410736 to
24c1227
Compare
support/environment.go
Outdated
| @@ -78,6 +79,10 @@ func GetRayImage() string { | |||
| return lookupEnvOrDefault(CodeFlareTestRayImage, RayImage) | |||
| } | |||
|
|
|||
| func GetRayAMDGpuImage() string { | |||
There was a problem hiding this comment.
I'd suggest to call it GetRayROCmImage, and maybe rename CodeFlareTestRayImage to CodeFlareTestRayCUDAImage.
jiripetrlik
force-pushed
the
RHOAIENG-11046
branch
from
24c1227 to
b5ba788
Compare
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: astefanutti, ChughShilpa, jiripetrlik The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-merge-bot
bot
merged commit 05a5ffd
into
project-codeflare:main
Issue link
What changes have been made
Verification steps
Checks