Provide NVIDIA T4 GPU support for snowex.uwhackweeks hub
#1784
Assignees
Comments
GeorgianaElena
moved this to Ready to work
in DEPRECATED Engineering and Product Backlog
yuvipanda
added a commit
to yuvipanda/pilot-hubs
that referenced
this issue
Oct 18, 2022
- The bug we reported upstream to eksctl has been fixed! So eksctl is now responsible for setting up the GPU driver, not us! eksctl-io/eksctl#5277. Yay for fixing things upstream! This woudl also mean that eksctl is responsible for keeping these versions up to date, and not us. - Based on pangeo-data/pangeo-docker-images#390 and many other discussions (linked to from there), NVidia T4s are now preferred over older K80s. We update the AWS GPU docs to recognize this. - Add PyTorch & Tensorflow images as options to the GPU profile here, so end users can choose! Fixes 2i2c-org#1784
|
We have some docs in https://infrastructure.2i2c.org/en/latest/howto/features/gpu.html?highlight=gpu on how to do this for AWS! |
yuvipanda
added a commit
to yuvipanda/pilot-hubs
that referenced
this issue
Oct 18, 2022
- The bug we reported upstream to eksctl has been fixed! So eksctl is now responsible for setting up the GPU driver, not us! eksctl-io/eksctl#5277. Yay for fixing things upstream! This woudl also mean that eksctl is responsible for keeping these versions up to date, and not us. We bump up the required eksctl version to account for this. - Based on pangeo-data/pangeo-docker-images#390 and many other discussions (linked to from there), NVidia T4s are now preferred over older K80s. We update the AWS GPU docs to recognize this. - Add PyTorch & Tensorflow images as options to the GPU profile here, so end users can choose! Fixes 2i2c-org#1784
|
@GeorgianaElena i took this on since I'm steeped in GPU support over the last few weeks. I've cleaned up docs to remove some steps, and deployed this as well. |
yuvipanda
added a commit
to yuvipanda/pilot-hubs
that referenced
this issue
Oct 18, 2022
- The bug we reported upstream to eksctl has been fixed! So eksctl is now responsible for setting up the GPU driver, not us! eksctl-io/eksctl#5277. Yay for fixing things upstream! This woudl also mean that eksctl is responsible for keeping these versions up to date, and not us. We bump up the required eksctl version to account for this. - Based on pangeo-data/pangeo-docker-images#390 and many other discussions (linked to from there), NVidia T4s are now preferred over older K80s. We update the AWS GPU docs to recognize this. - Add PyTorch & Tensorflow images as options to the GPU profile here, so end users can choose! Fixes 2i2c-org#1784
Repository owner
moved this from Ready to work
to Complete
in DEPRECATED Engineering and Product Backlog
Oct 18, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Context
Checout the support ticket https://2i2c.freshdesk.com/a/tickets/219 and #1765 (comment) for context.
Proposal
Offer access to nvidia T4 GPUs to the https://snowex.uwhackweeks.2i2c.cloud as a profile option (similar to m2lines & LEAP hubs in #1766)
Note:
The https://snowex.uwhackweeks.2i2c.cloud hub is deployed on AWS, where the T4 gpus are provided through EC@ G4 instances https://aws.amazon.com/ec2/instance-types/g4/
Updates and actions
No response
The text was updated successfully, but these errors were encountered: