What is the primary function of the kube-scheduler in Kubernetes?

What is the primary function of the kube-scheduler in Kubernetes?

• A. To monitor cluster health
• B. To manage networking
• C. To schedule applications to run on various nodes
• D. To store cluster state

Answer: (C)

Deciding on which node a Pod runs is the kube-scheduler's job. When a Pod is created without a bound node, the scheduler scans the cluster to find a suitable place for it. It filters out nodes that can’t fit the Pod—due to insufficient CPU/memory, unschedulable status, or taints that aren’t tolerable by the Pod. Then it scores the remaining nodes based on factors like current resource availability, data locality, affinity/anti-affinity rules, and Pod priority, and binds the Pod to the highest-scoring node. This orchestration ensures workloads are placed where they fit best while honoring constraints. The scheduler doesn’t monitor cluster health, manage networking, or store cluster state; those responsibilities lie with other components such as the API server/etcd and various controllers and networking plugins.

Deciding on which node a Pod runs is the kube-scheduler's job. When a Pod is created without a bound node, the scheduler scans the cluster to find a suitable place for it. It filters out nodes that can’t fit the Pod—due to insufficient CPU/memory, unschedulable status, or taints that aren’t tolerable by the Pod. Then it scores the remaining nodes based on factors like current resource availability, data locality, affinity/anti-affinity rules, and Pod priority, and binds the Pod to the highest-scoring node. This orchestration ensures workloads are placed where they fit best while honoring constraints. The scheduler doesn’t monitor cluster health, manage networking, or store cluster state; those responsibilities lie with other components such as the API server/etcd and various controllers and networking plugins.