Skip to content

Update GPU ec2 instance types in integration tests #8217

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

Merged
cheeseandcereal merged 1 commit into from
Feb 13, 2025
Merged

Update GPU ec2 instance types in integration tests #8217

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion integration/tests/crud/creategetdelete_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -924,7 +924,7 @@ var _ = Describe("(Integration) Create, Get, Scale & Delete", func() {
"--timeout=45m",
"--cluster", params.ClusterName,
"--nodes", "1",
"--instance-types", "p3.2xlarge,p3.8xlarge,g3s.xlarge,g4ad.xlarge,g4ad.2xlarge",
"--instance-types", "g6.xlarge,g6.2xlarge",
Copy link
Contributor

@bryantbiggs bryantbiggs Feb 13, 2025

Choose a reason for hiding this comment

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

should work but might run into insufficient capacity with this type FYI

if we go a bit larger to 4x, 8x, 16x - the price won't be too drastic given its run a short duration but there will be more available capacity (customers avoid these because they really need the GPUs, not the added vCPU/memory - so they all aim for the x, 2x size)
image

Copy link
Member Author

@cheeseandcereal cheeseandcereal Feb 13, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Yeah. I wanted to try using another instance type entirely, but they're either also old or gigantic (like the P5 instances start at 24xl).

We can try this for now, and then if there are issues adjust

"--node-private-networking",
"--node-zones", "us-west-2b,us-west-2c",
GPUMng,
Expand Down
Loading