Your GKE cluster currently runs on standard e2-medium nodes. Your data science team wants to deploy a new workload to the cluster that requires high-memory machines. You do not want to disrupt the existing workloads.

How should you accommodate this new workload?

A node pool is a group of nodes within a cluster that all have the same configuration (machine type, disk size, etc.). A single GKE cluster can have multiple node pools. If you need different hardware (like high-memory machines or GPUs) for specific workloads, the best practice is to add a new node pool to the existing cluster and use node selectors/taints to schedule the specific pods onto those nodes.