35 lines
1.1 KiB
YAML
35 lines
1.1 KiB
YAML
# services/ai-llm/gpu-reservation-job.yaml
|
|
apiVersion: batch/v1
|
|
kind: Job
|
|
metadata:
|
|
name: ollama-gpu-reserve-20260928
|
|
namespace: ai
|
|
spec:
|
|
backoffLimit: 1
|
|
activeDeadlineSeconds: 600
|
|
template:
|
|
metadata:
|
|
labels: {app: ollama-gpu-reservation}
|
|
spec:
|
|
nodeSelector:
|
|
kubernetes.io/hostname: titan-24
|
|
automountServiceAccountToken: false
|
|
hostPID: true
|
|
restartPolicy: Never
|
|
containers:
|
|
- name: reservation
|
|
image: debian@sha256:b6e2a152f22a40ff69d92cb397223c906017e1391a73c952b588e51af8883bf8
|
|
command: [/bin/bash, /scripts/gpu_pilot_reservation.sh, reserve]
|
|
securityContext: {privileged: true, runAsUser: 0}
|
|
resources:
|
|
requests: {cpu: 25m, memory: 64Mi}
|
|
limits: {cpu: 500m, memory: 128Mi}
|
|
volumeMounts:
|
|
- {name: host, mountPath: /host}
|
|
- {name: script, mountPath: /scripts, readOnly: true}
|
|
volumes:
|
|
- name: host
|
|
hostPath: {path: /, type: Directory}
|
|
- name: script
|
|
configMap: {name: ollama-gpu-reservation}
|