feature: integrate with kwok to simulate mock GPU/NPU nodes

JesseStutler commented 5 days ago

What type of PR is this?

/kind feature

What this PR does / why we need it:

related issue: #3829

verfication

Use ./create-fake-node.sh -n 10 -c 4 -m 8Gi -e volcano.sh/gpu-number=4,volcano.sh/gpu-memory=20 to create 10 fake nodes with 4 CPUs, 8Gi memories and extended resources with volcano.sh/gpu-number=4,volcano.sh/gpu-memory=20. After successfully creating these nodes, take one of them as an example:

Open deviceshare plugin and set the argument deviceshare.GPUNumberEnable enabled, and then create a fake deployment to create a pod requesting 1 volcano.sh/gpu-number and 1 volcano.sh/gpu-memory, successfully scheduled:

apiVersion: apps/v1
kind: Deployment
metadata:
name: fake-gpu-pod
namespace: default
spec:
replicas: 1
selector:
matchLabels:
  app: fake-gpu-pod
template:
metadata:
  labels:
    app: fake-gpu-pod
spec:
  schedulerName: volcano
  tolerations:
  - key: "kwok.x-k8s.io/node"
    operator: "Exists"
    effect: "NoSchedule"
  containers:
  - name: fake-container
    image: fake-image
    resources:
      limits:
        volcano.sh/gpu-number: 1
        volcano.sh/gpu-memory: 1

volcano-sh-bot commented 5 days ago

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: To complete the pull request process, please assign hwdef You can assign the PR to them by writing /assign @hwdef in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files: - **[OWNERS](https://github.com/volcano-sh/volcano/blob/master/OWNERS)** Approvers can indicate their approval by writing `/approve` in a comment Approvers can cancel approval by writing `/approve cancel` in a comment

JesseStutler commented 5 days ago

I found that under benchmark there is already a script to deploy kwok and fake nodes, maybe move these into benchmark/kwok is better.

volcano-sh / volcano