A development Kubernetes cluster deployed with NKP is having performance issues.

The NKPA course addresses performance issues in NKP clusters, such as high CPU and memory consumption on worker nodes, by recommending scaling options to distribute workloads more effectively. The Kubernetes cluster in the scenario has 4 workers, each with 8 vCPUs and 32 GB RAM, and is running out of resources. The most effective solution is toadd more workersto the cluster to increase overall capacity and balance the load, rather than modifying existing workers or reducing application replicas.

The correct action is toadd one more workerusing the command nkp scale nodepools ${NODEPOOL_NAME} --replicas=5 --cluster-name=${CLUSTER_NAME} -n ${CLUSTER_WORKSPACE} (Option D). This command scales the specified node pool to 5 replicas (from the current 4), adding one additional worker to the cluster. The new worker will have the same configuration (8 vCPUs, 32 GB RAM) as the existing workers, providing additional capacity to alleviate resource contention. The Nutanix Cloud Native (NCP-CN) 6.10 Study Guide states: “To address performance issues due to resource exhaustion in an NKP cluster, scale out by adding workers using nkp scale nodepools --replicas --cluster-name -n .”

Incorrect Options:

A. Call tech support to investigate: While support can help, the issue is clearly resource exhaustion, which can be resolved by scaling the cluster—a task the Platform Engineer can perform directly.

B. Ask developers to lower application replicas: Reducing replicas may decrease resource usage but could impact application availability or performance, which is not ideal for a development cluster.

C. Add more CPU and memory to workers with nkp scale --cpu 16 --memory 64: The nkp scale command does not support --cpu and --memory flags for resizing existing workers. Resizing VMs typically requires updating the node pool configuration and replacing nodes, which is more complex than adding new workers.