Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Test with GPU and m+t4 machines #533

Merged
merged 42 commits into from
Apr 27, 2022
Merged

Test with GPU and m+t4 machines #533

merged 42 commits into from
Apr 27, 2022

Conversation

0x2b3bfa0
Copy link
Member

@0x2b3bfa0 0x2b3bfa0 commented Apr 25, 2022 •
edited
Loading

This pull request replaces the official NVIDIA NGC images (#384, #410) by the ones offered by cloud providers. Now image = "nvidia" uses the images described in #410 (comment).

It also removes the fine–grained GPU selectors for Kubernetes, because AKS only provides accelerator=nvidia with the node labels, and it's not possible to set that label when creating the cluster.

A solution we can implement, post–release and once tests are green, is weighted affinity, prioritizing the exact GPU model but tolerating any NVIDIA GPU device to avoid errors.

⚠️ Warning: testing k8s with NVIDIA K80 until Azure lifts the quota restrictions; then, nodes should be moved to Standard_NC4as_T4_v3

@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 08:53 Inactive
@restyled-io restyled-io bot mentioned this pull request Apr 25, 2022
Merged
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 08:59 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 09:00 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 09:12 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 09:12 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 25, 2022 09:13 Inactive
casperdcl
casperdcl reviewed Apr 27, 2022
View reviewed changes
iterative/resource_task.go Show resolved Hide resolved
task/k8s/resources/resource_job.go Show resolved Hide resolved
task/k8s/resources/resource_job.go Outdated Show resolved Hide resolved
@0x2b3bfa0 @casperdcl
Update task/k8s/resources/resource_job.go
c7f3041 
Co-authored-by: Casper da Costa-Luis <casper.dcl@physics.org>
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 27, 2022 10:59 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 27, 2022 11:00 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 27, 2022 11:00 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 27, 2022 11:00 Inactive
@0x2b3bfa0 0x2b3bfa0 temporarily deployed to automatic April 27, 2022 11:00 Inactive
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:29 Inactive
@casperdcl casperdcl self-assigned this Apr 27, 2022
casperdcl
casperdcl reviewed Apr 27, 2022
View reviewed changes
docs/resources/task.md Outdated Show resolved Hide resolved
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:37 Inactive
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:37 Inactive
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:37 Inactive
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:37 Inactive
@casperdcl casperdcl temporarily deployed to automatic April 27, 2022 11:37 Inactive
@0x2b3bfa0
Copy link
Member Author

@iterative/cml, ready for merging? 🚀

DavidGOrtega
DavidGOrtega approved these changes Apr 27, 2022
View reviewed changes
Copy link
Contributor

@DavidGOrtega DavidGOrtega left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

I do not have tested it nor know if the images are OK. Everything looks fine to me so far

@0x2b3bfa0 0x2b3bfa0 merged commit 672acef into master Apr 27, 2022
@0x2b3bfa0 0x2b3bfa0 deleted the test-m-t4 branch April 27, 2022 13:05
This was referenced Apr 27, 2022
Open
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@casperdcl casperdcl casperdcl left review comments

@DavidGOrtega DavidGOrtega DavidGOrtega approved these changes

Assignees

@casperdcl casperdcl

@0x2b3bfa0 0x2b3bfa0

Labels
cloud-common Applies to every cloud vendor p1-important High priority resource-task iterative_task TF resource testing Unit tests & debugging
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@0x2b3bfa0 @DavidGOrtega @casperdcl @restyled-commits