A. By selecting a managed node pool, a GPU shape, and a GPU image, and ensuring that the CUDA libraries for different GPUs are pre-installed on the worker nodes

B. By selecting a managed node pool, a GPU shape, and a GPU image that has CUDA libraries pre-installed, and specifying the number of GPU resources required in the pod spec

C. By selecting a managed node pool, a GPU shape, and a GPU image, and specifying the number of CPU resources required in the pod spec

D. By selecting a virtual node pool, a GPU shape, and a GPU image, and ensuring that the CUDA libraries are included in the application container

A. Build pipelines

B. Integration pipelines

C. Development pipelines

D. Deployment pipelines

A. By disabling access to the Container Engine for Kubernetes cluster

B. By manually inspecting each image before deployment

C. By encrypting the images using a custom encryption algorithm

D. By configuring an image verification policy for the cluster

A. Container Instances require the team to perform provisioning, patching, and ongoing management of servers.

B. Container Instances provide a serverless compute service for running containers, eliminating the need for server management.

C. Billing for Container Instances is based on the allocated CPU and memory resources for each container instance.

D. The underlying infrastructure for Container Instances is fully managed and hardened by OCI, ensuring reliability and security.

A. Service Connector

B. Actions

C. Management Agent Cloud Service

D. Auth Token

E. Rule Conditions

A. Generating an API signing key pair is not required while setting up cluster access using local machine if the public key is not already uploaded in the console.

B. When a cluster's Kubernetes API endpoint has a public IP address, you can access the cluster in Cloud Shell by setting up a kubeconfig file.

C. The only available option when a cluster's Kubernetes API endpoint has a public IP address is to control the cluster locally using kubectl and the Kubernetes Dashboard.

D. To access the cluster using kubectl, you have to set up a Kubernetes manifest file for the cluster. The kubeconfig file by default is named config and stored in the $HOME/.manifest directory.

E. To access the cluster using kubectl, you have to set up a Kubernetes configuration file for the cluster. The kubeconfig file by default is named config and stored in the $HOME/.kube directory.