Adding GFD

danmanners · Nov 11, 2023 · 1890f43 · 1890f43
1 parent 2459b4e
commit 1890f43
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 0 deletions.
diff --git a/manifests/workloads/applicationset.yaml b/manifests/workloads/applicationset.yaml
@@ -33,6 +33,9 @@ spec:
       - appName: node-feature-discovery
         namespace: kube-system
         targetRev: main
+      - appName: nvidia-device-plugin
+        namespace: kube-system
+        targetRev: main
       - appName: qemu-guest-agent
         namespace: kube-system
         targetRev: main

diff --git a/manifests/workloads/nvidia/kustomization.yaml b/manifests/workloads/nvidia/kustomization.yaml
@@ -10,3 +10,28 @@ helmCharts:
   version: 0.14.2
   valuesInline:
     runtimeClassName: nvidia
+    gfd:
+      enabled: true
+    affinity:
+      nodeAffinity:
+        requiredDuringSchedulingIgnoredDuringExecution:
+          nodeSelectorTerms:
+          - matchExpressions:
+            # On discrete-GPU based systems NFD adds the following label where 10de is the NVIDIA PCI vendor ID
+            - key: feature.node.kubernetes.io/pci-10de.present
+              operator: In
+              values:
+              - "true"
+          - matchExpressions:
+            # On some Tegra-based systems NFD detects the CPU vendor ID as NVIDIA
+            - key: feature.node.kubernetes.io/cpu-model.vendor_id
+              operator: In
+              values:
+              - "NVIDIA"
+          - matchExpressions:
+            # We allow a GPU deployment to be forced by setting the following label to "true"
+            - key: "nvidia.com/gpu.present"
+              operator: In
+              values:
+              - "true"
+