Skip to content

Added functionality in show_gpus to have an option to show all regions' pricing for an accelerator #2892

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.

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
merged 11 commits into from
Dec 30, 2023

Conversation

vaibagra
Copy link
Contributor

@vaibagra vaibagra commented Dec 22, 2023

This solves #2863

show_gpus can now accept a new flag --all-regions, which displays all the regions for a given accelerator. It can be used as follows.

sky show-gpus <acclerator> --all-regions

Additionally it can take a cloud argument as well.

sky show-gpus <acclerator> --cloud <cloud> --all-regions

I compared the outputs against the catalog.

Tested (run the relevant ones):

  • Code formatting: bash format.sh
  • Checking all regions exist.

Copy link
Collaborator

@romilbhardwaj romilbhardwaj left a comment

Choose a reason for hiding this comment

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

Thanks @Vaibhav2001! Seems to work nicely. Left some comments.

Copy link
Collaborator

@romilbhardwaj romilbhardwaj left a comment

Choose a reason for hiding this comment

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

Thanks @Vaibhav2001! Tested with clouds + a Kubernetes T4 cluster, seems to be working well. LGTM.

@romilbhardwaj romilbhardwaj merged commit 6c30d79 into skypilot-org:master Dec 30, 2023
@concretevitamin
Copy link
Member

Awesome work @Vaibhav2001.

For the same gpu:count, it may be good to sort by (HOURLY_PRICE, HOURLY_SPOT_PRICE, REGION). The current behavior appears to be sorting by REGION.

» sky show-gpus --all-regions A100 --cloud gcp
*NOTE*: for most GCP accelerators, INSTANCE_TYPE == (attachable) means the host VM's cost is not included.

GPU   QTY  CLOUD  INSTANCE_TYPE   DEVICE_MEM  vCPUs  HOST_MEM  HOURLY_PRICE  HOURLY_SPOT_PRICE  REGION
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.841       $ 1.536            asia-northeast1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.841       $ 1.536            asia-northeast3
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.841       $ 1.536            asia-southeast1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.673       $ 1.176            europe-west4
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.967       $ 1.587            me-west1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.673       $ 1.469            us-central1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.673       $ 1.469            us-east1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.673       $ 1.469            us-west1
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 4.176       $ 1.670            us-west3
A100  1    GCP    a2-highgpu-1g   -           12     85GB      $ 3.841       $ 1.536            us-west4

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants