feat(gpu): Change from nvidia-gpu-operator to nvidia-device-plugin

The simpler nvidia-device-plugin chart covers my needs and is easier to
get working.

infra/gpu-operator/cm-device-plugin-time-slicing.yaml

@@ -1,15 +0,0 @@
-apiVersion: v1
-kind: ConfigMap
-metadata:
-  name: device-plugin-time-slicing
-  namespace: gpu-operator
-data:
-  any: |-
-    version: v1
-    flags:
-      migStrategy: none
-    sharing:
-      timeSlicing:
-        resources:
-        - name: nvidia.com/gpu
-          replicas: 10

infra/gpu-operator/kustomization.yaml

@@ -1,15 +0,0 @@
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-namespace: gpu-operator
-
-resources:
-  - namespace.yaml
-  - cm-device-plugin-time-slicing.yaml
-
-helmCharts:
-  - name: gpu-operator
-    repo: https://helm.ngc.nvidia.com/nvidia
-    version: 23.9.0
-    releaseName: "gpu-operator"
-    includeCRDs: true
-    valuesFile: values.yaml

infra/gpu-operator/values.yaml

@@ -1,13 +0,0 @@
-driver:
-  enabled: false
-toolkit:
-  enabled: false
-
-devicePlugin:
-  config:
-    name: device-plugin-time-slicing
-    default: any
-
-#cdi:
-#  enabled: true
-#  default: true

infra/nvidia-device-plugin/cm-time-slicing.yaml

@@ -0,0 +1,12 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: time-slicing
+data:
+  default: |-
+    version: v1
+    sharing:
+      timeSlicing:
+        resources:
+          - name: nvidia.com/gpu
+            replicas: 10

infra/nvidia-device-plugin/kustomization.yaml

@@ -0,0 +1,16 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+namespace: nvidia-device-plugin
+
+resources:
+  - namespace.yaml
+  - cm-time-slicing.yaml
+
+helmCharts:
+  - name: nvidia-device-plugin
+    repo: https://nvidia.github.io/k8s-device-plugin
+    version: 0.14.2
+    releaseName: "nvidia-device-plugin"
+    namespace: nvidia-device-plugin
+    includeCRDs: true
+    valuesFile: values.yaml

infra/nvidia-device-plugin/namespace.yaml

@@ -1,4 +1,4 @@
 apiVersion: v1
 kind: Namespace
 metadata:
-  name: gpu-operator
+  name: nvidia-device-plugin

infra/nvidia-device-plugin/values.yaml

@@ -0,0 +1,3 @@
+config:
+  name: time-slicing
+  default: default

infra/project.yaml

@@ -24,6 +24,8 @@ spec:
       server: '*'
     - namespace: 'monitoring'
       server: '*'
+    - namespace: 'nvidia-device-plugin'
+      server: '*'
     - namespace: 'pi-hole'
       server: '*'
     - namespace: 'net-aux'
@@ -34,4 +36,4 @@ spec:
       server: '*'
   clusterResourceWhitelist:
     - group: '*'
-      kind: '*'
+      kind: '*'