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.

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

Suggestion update to the new instaslice api structure #450

Open
mamy-CS opened this issue Feb 20, 2025 · 0 comments
Open

Suggestion update to the new instaslice api structure #450

mamy-CS opened this issue Feb 20, 2025 · 0 comments

Comments

@mamy-CS
Copy link
Contributor

mamy-CS commented Feb 20, 2025

Add hierarchy to Node Gpus as such and move under Spec, Node Resources:
Current:

Status:
  Node Resources:
    Node Gpus:
       Gpu Memory: 40Gi
       Gpu Name:  NVIDIA A100-PCIE-40GB
       Gpu Uuid:  GPU-8d042338-e67f-9c48-92b4-5b55c7e5133c
       Gpu Memory: 40Gi
       Gpu Name:  NVIDIA A100-PCIE-40GB
       Gpu Uuid:  GPU-31cfe05c-ed13-cd17-d7aa-c63db5108c24

Update:

Spec:
  Node Resources:
    Node Gpus:
       Gpu Uuid:  GPU-8d042338-e67f-9c48-92b4-5b55c7e5133c
          Gpu Memory: 40Gi
          Gpu Name:  NVIDIA A100-PCIE-40GB
       Gpu Uuid:  GPU-31cfe05c-ed13-cd17-d7aa-c63db5108c24
          Gpu Memory: 40Gi
          Gpu Name:  NVIDIA A100-PCIE-40GB

Also:
Would it make sense to add Pod Allocation Requests under Status to avoid confusion since Pod Allocation Results is also under Status? To keep all pod info in the same area. I believe it would simplify some code as well. As such maybe?

Status:
  Pod Allocation Requests:
    db18292b-4ddf-44c2-ba6d-797c407191ab:
      Pod Ref:
        Kind:       Pod
        Name:       emulator-pod-1
        Namespace:  default
        UID:        db18292b-4ddf-44c2-ba6d-797c407191ab
      Profile:      1g.5gb
      Resources:
        Requests:
          Cpu:     0
          Memory:  0
  Pod Allocation Results:
      db18292b-4ddf-44c2-ba6d-797c407191ab:
        Allocation Status:
          Allocation Status Controller:  ungated
          Allocation Status Daemonset:   created
        Config Map Resource Identifier:  3391ab2f-24a9-4a4e-98c4-9973c5a06cbf
        Gpu UUID:                        GPU-31cfe05c-ed13-cd17-d7aa-c63db5108c24
        Mig Placement:
          Size:    1
          Start:   0
        Nodename:  kind-e2e-control-plane

or maybe add it under the same pod?

To put it all together what I am suggesting:

Spec:
  Node Resources:
    Mig Placement:
      1g.10gb:
        Ci Profile Id:  9
        Gi Profile Id:  9
        Placements:
          Size:   2
          Start:  0
          Size:   2
          Start:  2
          Size:   2
          Start:  4
          Size:   2
          Start:  6
      1g.5gb:
        Ci Profile Id:  0
        Gi Profile Id:  0
        Placements:
          Size:   1
          Start:  0
          Size:   1
          Start:  1
          Size:   1
          Start:  2
          Size:   1
          Start:  3
          Size:   1
          Start:  4
          Size:   1
          Start:  5
          Size:   1
          Start:  6
      1g.5gb+me:
        Ci Profile Id:  7
        Gi Profile Id:  7
        Placements:
          Size:   1
          Start:  0
          Size:   1
          Start:  1
          Size:   1
          Start:  2
          Size:   1
          Start:  3
          Size:   1
          Start:  4
          Size:   1
          Start:  5
          Size:   1
          Start:  6
      2g.10gb:
        Ci Profile Id:  1
        Gi Profile Id:  1
        Placements:
          Size:   2
          Start:  0
          Size:   2
          Start:  2
          Size:   2
          Start:  4
      3g.20gb:
        Ci Profile Id:  2
        Gi Profile Id:  2
        Placements:
          Size:   4
          Start:  0
          Size:   4
          Start:  4
      4g.20gb:
        Ci Profile Id:  3
        Gi Profile Id:  3
        Placements:
          Size:   4
          Start:  0
      7g.40gb:
        Ci Profile Id:  4
        Gi Profile Id:  4
        Placements:
          Size:   8
          Start:  0
    Node Gpus:
       Gpu Uuid:  GPU-8d042338-e67f-9c48-92b4-5b55c7e5133c
          Gpu Memory: 40Gi
          Gpu Name:  NVIDIA A100-PCIE-40GB
       Gpu Uuid:  GPU-31cfe05c-ed13-cd17-d7aa-c63db5108c24
          Gpu Memory: 40Gi
          Gpu Name:  NVIDIA A100-PCIE-40GB
    Node Resources:
      Cpu:     72
      Memory:  1G
Status:
  Pod Allocation Requests:
    db18292b-4ddf-44c2-ba6d-797c407191ab:
      Pod Ref:
        Kind:       Pod
        Name:       emulator-pod-1
        Namespace:  default
        UID:        db18292b-4ddf-44c2-ba6d-797c407191ab
      Profile:      1g.5gb
      Resources:
        Requests:
          Cpu:     0
          Memory:  0
  Pod Allocation Results:
    db18292b-4ddf-44c2-ba6d-797c407191ab:
      Allocation Status:
        Allocation Status Controller:  ungated
        Allocation Status Daemonset:   created
      Config Map Resource Identifier:  3391ab2f-24a9-4a4e-98c4-9973c5a06cbf
      Gpu UUID:                        GPU-31cfe05c-ed13-cd17-d7aa-c63db5108c24
      Mig Placement:
        Size:    1
        Start:   0
      Nodename:  kind-e2e-control-plane
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

No branches or pull requests

1 participant