kubernetes training on optocycle clusterHEADmain

1 files changed, 70 insertions, 0 deletions

diff --git a/pod.yaml b/pod.yaml
new file mode 100644
index 0000000..a415c04
--- /dev/null
+++ b/pod.yaml
@@ -0,0 +1,70 @@
+# a pod for testing the torch implementation of nms on a gpu
+apiVersion: v1
+kind: Pod
+metadata:
+  name: swr-asr
+  labels:
+    app: swr-asr
+spec:
+  initContainers:
+    - name: swr-asr-init
+      image: pherkel/swr2-asr
+      imagePullPolicy: Always
+      command: ["/bin/sh", "-c", "cp -R /app/* /tmp/"]
+      volumeMounts:
+        - name: swr-asr-vol
+          mountPath: /tmp
+
+      securityContext:
+        allowPrivilegeEscalation: false
+        capabilities:
+          drop: ["ALL"]
+
+
+  containers:
+    - name: swr-asr
+      image: pherkel/swr2-asr
+      imagePullPolicy: Always
+      # command: ["/bin/bash", "-c", "while sleep 1000; do :; done"]
+
+      volumeMounts:
+        - name: swr-asr-vol
+          mountPath: /app
+
+        # workaround for increasing worker shared memory size
+        - name: shm-vol
+          mountPath: /dev/shm
+
+      resources:
+        requests:
+          nvidia.com/gpu: "1"
+          memory: "8Gi"
+          cpu: "4"
+        limits:
+          nvidia.com/gpu: "1"
+          memory: "8Gi"
+          cpu: "4"
+
+      securityContext:
+        allowPrivilegeEscalation: false
+        capabilities:
+          drop: ["ALL"]
+
+  volumes:
+    - name: swr-asr-vol
+      persistentVolumeClaim:
+        claimName: swr-asr-vol
+    - name: shm-vol
+      emptyDir:
+        medium: Memory
+
+  securityContext:
+    runAsNonRoot: true
+    runAsUser: 1000
+    runAsGroup: 1000
+    fsGroup: 1000
+    seccompProfile:
+      type: RuntimeDefault
+
+  nodeSelector:
+    optocycle.com/infrastructure-provider: ocs