Hi NV team,
Please let me confirm the following question.
Would you please teach me the way to run the RAPIDS on K8s environments?
Now I created the simple yaml file to deploy the RAPIDS but it was happened the “CrashLoopBackOff”.
If possible, would you please check my below yaml file?
apiVersion: v1
kind: Pod
metadata:
name: rapids
labels:
app: rapids
spec:
containers:
- name: rapids
image: nvcr.io/nvidia/rapidsai/rapidsai:cuda10.0-runtime-ubuntu18.04
imagePullPolicy: IfNotPresent
ports:- name: jupyter
containerPort: 8888
protocol: TCP - name: backporta
containerPort: 8787
protocol: TCP - name: backportb
containerPort: 8786
protocol: TCP
resources:
limits:
nvidia.com/gpu: 1
- name: jupyter
kind: Service
apiVersion: v1
metadata:
name: rapids-service
spec:
type: ClusterIP
externalIPs:
- 10.13.20.40
ports:
- name: “jupyter”
protocol: “TCP”
port: 8889
targetPort: jupyter
- name: “backporta”
protocol: “TCP”
port: 8787
targetPort: backporta
- name: “backportb”
protocol: “TCP”
port: 8786
targetPort: backportb
selector:
app: rapids
Best regards.
Kaka