Refresh CUDA example images

docs/airgap/mirror-docker-images.md

@@ -42,9 +42,8 @@ Then, for each image you want to download, you should pull the image from the re
 In this example, we're saving all our Docker images to `/tmp/images`:
 
 ```
-$ docker pull nvidia/cuda:11.1-devel-ubuntu20.04
-$ docker save -o /tmp/images/nvidia-cuda-11.1-devel-ubuntu20.04.tar nvidia/cuda:11.1-devel-ubuntu20.04 
->>>>>>> e1d0a775 (Airgap documentation update.)
+$ docker pull nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
+$ docker save -o /tmp/images/nvidia-cuda-12.4.1-base-ubuntu22.04.tar nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
 ```
 
 Additionally, you should download and save the [`registry` image](https://hub.docker.com/_/registry) so that you can deploy a local registry on the offline network.
@@ -76,13 +75,8 @@ Additionally, we assume that the `registry` image was included when you transfer
 
 Load the registry image into the Docker image cache of your container registry host:
 
-<<<<<<< HEAD
 ```bash
-docker load < /tmp/images/registry-2.7.tar
-=======
-```
-$ docker load -i /tmp/images/registry-2.7.tar
->>>>>>> e1d0a775 (Airgap documentation update.)
+$ docker load -i /tmp/images/registry-3.1.1.tar
 ```
 
 Then create a Docker volume to store your container images:
@@ -99,7 +93,7 @@ docker run -d \
     --restart=always \
     --name registry \
     -v registry-images:/var/lib/registry \
-    registry:2.7
+    registry:3.1.1
 ```
 
 ## Configuring your hosts to use the offline container registry
@@ -121,7 +115,7 @@ docker_insecure_registries:
 Once your registry is running and you've configured your hosts to access it, you can load additional images and push them to the offline registry:
 
 ```bash
-docker load < /tmp/images/nvidia-cuda-11.1-devel-ubuntu20.04.tar
-docker tag nvidia/cuda:11.1-devel-ubuntu20.04 registry-host:5000/nvidia/cuda:11.1-devel-ubuntu20.04
-docker push registry-host:5000/nvidia/cuda:11.1-devel-ubuntu20.04
+docker load -i /tmp/images/nvidia-cuda-12.4.1-base-ubuntu22.04.tar
+docker tag nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04 registry-host:5000/nvidia/cuda:12.4.1-base-ubuntu22.04
+docker push registry-host:5000/nvidia/cuda:12.4.1-base-ubuntu22.04
 ```

docs/airgap/ngc-ready.md

@@ -37,9 +37,9 @@ For instructions on setting up an HTTP mirror, see the [doc on HTTP mirrors](./m
 
 Container images are only needed if you want to run the tests built into the playbook:
 
-- nvcr.io/nvidia/cuda:10.1-base-ubuntu18.04
-- nvcr.io/nvidia/pytorch:18.10-py3
-- nvcr.io/nvidia/tensorflow:18.10-py3
+- nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
+- nvcr.io/nvidia/pytorch:24.04-py3
+- nvcr.io/nvidia/tensorflow:24.04-tf2-py3
 
 For instructions on setting up a Docker registry mirror, see the [doc on Docker mirrors](./mirror-docker-images.md).
 
@@ -63,9 +63,9 @@ For instructions on setting up an HTTP mirror, see the [doc on HTTP mirrors](./m
 
 Container images (how to mirror) are only needed if you want to run the tests built into the playbook:
 
-- nvcr.io/nvidia/cuda:10.1-base-ubuntu18.04
-- nvcr.io/nvidia/pytorch:18.10-py3
-- nvcr.io/nvidia/tensorflow:18.10-py3
+- nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
+- nvcr.io/nvidia/pytorch:24.04-py3
+- nvcr.io/nvidia/tensorflow:24.04-tf2-py3
 
 For instructions on setting up a Docker registry mirror, see the [doc on Docker mirrors](./mirror-docker-images.md).
 
@@ -182,9 +182,9 @@ dcgm_rpm_package: "/path/to/datacenter-gpu-manager.rpm"
 If running the container tests as part of the NGC-Ready playbook, set the following variables in your DeepOps configuration:
 
 ```bash
-ngc_ready_cuda_container: "<your-container-registry>/nvidia/cuda:10.1-base-ubuntu18.04"
-ngc_ready_pytorch: "<your-container-registry>/nvidia/pytorch:18.10-py3"
-ngc_ready_tensorflow: "<your-container-registry>/nvidia/tensorflow:18.10-py3"
+ngc_ready_cuda_container: "<your-container-registry>/nvidia/cuda:12.4.1-base-ubuntu22.04"
+ngc_ready_pytorch: "<your-container-registry>/nvidia/pytorch:24.04-py3"
+ngc_ready_tensorflow: "<your-container-registry>/nvidia/tensorflow:24.04-tf2-py3"
 ```
 
 ## Running the NGC-Ready playbook

virtual/scripts/setup_k8s.sh

@@ -40,7 +40,7 @@ source "${VIRT_DIR}/k8s_environment.sh"
 # Verify that the cluster is up
 file ${K8S_CONFIG_DIR}/artifacts/kubectl && chmod +x ${K8S_CONFIG_DIR}/artifacts/kubectl
 kubectl get nodes
-#kubectl run gpu-test --rm -t -i --restart=Never --image=nvcr.io/nvidia/cuda:10.1-base-ubuntu18.04 --limits=nvidia.com/gpu=1 -- nvidia-smi
+#kubectl run gpu-test --rm -t -i --restart=Never --image=nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04 --limits=nvidia.com/gpu=1 -- nvidia-smi
 
 # Install helm
 "${ROOT_DIR}/scripts/k8s/install_helm.sh"

workloads/examples/k8s/cluster-gpu-test-job.yml

@@ -10,7 +10,7 @@ spec:
     spec:
       containers:
         - name: cluster-gpu-tests
-          image: nvcr.io/nvidia/cuda:9.0-base 
+          image: nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
           command: ["/bin/bash","-c","nvidia-smi && sleep 10"]
           args:
           resources:

workloads/examples/k8s/gpu-test-job.yml

@@ -5,11 +5,10 @@ metadata:
 spec:
   containers:
     - name: cuda-container
-      image: nvcr.io/nvidia/cuda:10.0-devel
+      image: nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04
       command: ["sleep", "6000"]
       args:
       resources:
         limits:
           nvidia.com/gpu: 1
   restartPolicy: Never
-

workloads/jenkins/scripts/run-gpu-job.sh

@@ -32,7 +32,7 @@ fi
 
 # Occassionally this gpu-test fails and/or hangs. To ease debugging of this we run a describe several seconds into the launch.
 sleep 10 && kubectl describe pods gpu-test &
-timeout 300 kubectl run gpu-test --rm -t -i --restart=Never --image=nvcr.io/nvidia/cuda:10.1-base-ubuntu18.04 --limits=nvidia.com/gpu=1 -- nvidia-smi
+timeout 300 kubectl run gpu-test --rm -t -i --restart=Never --image=nvcr.io/nvidia/cuda:12.4.1-base-ubuntu22.04 --limits=nvidia.com/gpu=1 -- nvidia-smi
 
 # Run multi-GPU test
 if [ ${DEEPOPS_FULL_INSTALL} ]; then

workloads/jenkins/scripts/test-slurm-enroot-job.sh

@@ -9,5 +9,5 @@ ssh -v \
 	-i "${HOME}/.ssh/id_rsa" \
 	"10.0.0.5${GPU01}" \
 	srun -N1 -G1 \
-	--container-image="nvcr.io#nvidia/cuda:10.2-base-ubuntu18.04" \
+	--container-image="nvcr.io#nvidia/cuda:12.4.1-base-ubuntu22.04" \
 	nvidia-smi -L