gcloud-lab/apps/base/customer1/news_bot/deepseek-summarizer.yaml

72 lines
2 KiB
YAML
Raw Normal View History

apiVersion: batch/v1
kind: CronJob
metadata:
name: deepseek-analyst
namespace: customer1
spec:
schedule: "15 * * * *"
jobTemplate:
spec:
template:
spec:
2026-03-25 02:01:26 +00:00
nodeSelector:
cloud.google.com/gke-accelerator: nvidia-h100-80gb
tolerations:
- key: "nvidia.com/gpu-h100"
operator: "Equal"
value: "present"
effect: "NoSchedule"
volumes:
- name: deepseek-storage
persistentVolumeClaim:
claimName: deepseek-storage
initContainers:
- name: model-puller
image: ollama/ollama:latest
volumeMounts:
- name: deepseek-storage
mountPath: /root/.ollama
command: ["/bin/sh", "-c"]
args: ["ollama serve & sleep 5 && ollama pull deepseek-r1:70b"]
- name: ollama-sidecar
image: ollama/ollama:latest
restartPolicy: Always
env:
- name: OLLAMA_DEBUG
value: "2"
- name: OLLAMA_LOG_LEVEL
value: "DEBUG"
ports:
- containerPort: 11434
volumeMounts:
- name: deepseek-storage
mountPath: /root/.ollama
resources:
limits:
nvidia.com/gpu: 1
memory: "85Gi"
requests:
nvidia.com/gpu: "1"
2026-03-25 02:01:26 +00:00
memory: "80Gi"
containers:
- name: analyst
image: siriussec/summarizerlocal:3.5
imagePullPolicy: Always
env:
- name: DB_PASSWORD
valueFrom:
secretKeyRef:
name: news-app-password
key: password
envFrom:
- configMapRef:
name: deepseek-config
restartPolicy: OnFailure