From 23e5148814f260520f01017ccf6e78241a7a5a91 Mon Sep 17 00:00:00 2001
From: AB-UK <ab@hexor.cy>
Date: Fri, 6 Mar 2026 23:24:40 +0000
Subject: [PATCH] feat: Add NVIDIA GPU config and update Ollama to use it

---
 k8s/apps/ollama/ollama-values.yaml |  7 ++++++-
 k8s/core/gpu/app.yaml              | 20 ++++++++++++++++++++
 k8s/core/gpu/kustomization.yaml    | 15 +++++++++++++++
 k8s/core/gpu/runtime-class.yaml    |  5 +++++
 k8s/core/gpu/values.yaml           | 21 +++++++++++++++++++++
 5 files changed, 67 insertions(+), 1 deletion(-)
 create mode 100644 k8s/core/gpu/app.yaml
 create mode 100644 k8s/core/gpu/kustomization.yaml
 create mode 100644 k8s/core/gpu/runtime-class.yaml
 create mode 100644 k8s/core/gpu/values.yaml

diff --git a/k8s/apps/ollama/ollama-values.yaml b/k8s/apps/ollama/ollama-values.yaml
index 4d370b7..57899c4 100644
--- a/k8s/apps/ollama/ollama-values.yaml
+++ b/k8s/apps/ollama/ollama-values.yaml
@@ -3,6 +3,11 @@ image:
   pullPolicy: Always
   tag: "latest"
 nodeSelector:
-  kubernetes.io/hostname: master.tail2fe2d.ts.net
+  kubernetes.io/hostname: uk-desktop.tail2fe2d.ts.net
 ingress:
     enabled: false
+ollama:
+  gpu:
+    enabled: true
+    type: 'nvidia'
+    number: 1
diff --git a/k8s/core/gpu/app.yaml b/k8s/core/gpu/app.yaml
new file mode 100644
index 0000000..dff3718
--- /dev/null
+++ b/k8s/core/gpu/app.yaml
@@ -0,0 +1,20 @@
+apiVersion: argoproj.io/v1alpha1
+kind: Application
+metadata:
+  name: gpu-system
+  namespace: argocd
+spec:
+  project: core
+  destination:
+    namespace: gpu-system
+    server: https://kubernetes.default.svc
+  source:
+    repoURL: ssh://git@gt.hexor.cy:30022/ab/homelab.git
+    targetRevision: HEAD
+    path: k8s/core/gpu
+  syncPolicy:
+    automated:
+      selfHeal: true
+      prune: true
+    syncOptions:
+      - CreateNamespace=true
diff --git a/k8s/core/gpu/kustomization.yaml b/k8s/core/gpu/kustomization.yaml
new file mode 100644
index 0000000..45a573e
--- /dev/null
+++ b/k8s/core/gpu/kustomization.yaml
@@ -0,0 +1,15 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - app.yaml
+  - runtime-class.yaml
+
+helmCharts:
+  - name: nvidia-device-plugin
+    repo: https://nvidia.github.io/k8s-device-plugin
+    version: 0.17.0
+    releaseName: nvidia-device-plugin
+    namespace: gpu-system
+    valuesFile: values.yaml
+    includeCRDs: true
diff --git a/k8s/core/gpu/runtime-class.yaml b/k8s/core/gpu/runtime-class.yaml
new file mode 100644
index 0000000..c26bd6d
--- /dev/null
+++ b/k8s/core/gpu/runtime-class.yaml
@@ -0,0 +1,5 @@
+apiVersion: node.k8s.io/v1
+kind: RuntimeClass
+metadata:
+  name: nvidia
+handler: nvidia
diff --git a/k8s/core/gpu/values.yaml b/k8s/core/gpu/values.yaml
new file mode 100644
index 0000000..35d603d
--- /dev/null
+++ b/k8s/core/gpu/values.yaml
@@ -0,0 +1,21 @@
+nodeSelector:
+  kubernetes.io/hostname: uk-desktop.tail2fe2d.ts.net
+
+runtimeClassName: nvidia
+
+setAsDefault: false
+
+config:
+  name: "nvidia-plugin-config"
+  map:
+    # use device driver mode instead of dev/nvmem
+    default: |-
+      version: v1
+      flags:
+        migStrategy: none
+        failOnInitError: true
+        nvidiaDriverRoot: "/"
+        plugin:
+          passDeviceSpecs: false
+          deviceListStrategy: envvar
+          deviceIDStrategy: uuid