Running a Job on the Cluster

1. Create job-hl.yaml file. The job will run the hl-smi tool to display a summary table of the Gaudi devices. The following is an example of job-hl.yaml file:

   apiVersion: batch/v1
   kind: Job
   metadata:
       name: job-hl
   spec:
     template:
       metadata:
           labels:
               app: job-hl
       spec:
           restartPolicy: Never
           hostIPC: true
           containers:
           - name: job-hl
             image: vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.0:latest
             command: ["hl-smi"]
             workingDir: /home
             securityContext:
               capabilities:
                 add: ["SYS_NICE"]
             resources:
               limits:
                 habana.ai/gaudi: 8
                 hugepages-2Mi: "42000Mi"
                 memory: 690Gi
                 vpc.amazonaws.com/efa: 4
               requests:
                 habana.ai/gaudi: 8
                 hugepages-2Mi: "42000Mi"
                 memory: 690Gi
                 vpc.amazonaws.com/efa: 4
   

2. Run the job:

   kubectl apply -f job-hl.yaml
   

3. Check the job status:

   kubectl get pods -A
   

4. Retrieve the name of the created pod and run the following command to see the results:

   kubectl logs <pod-name>