kube-scheduler

Component watching newly created pods and assign them a node to run on

When a new pod is created, determines which node the pod will be run on according to different factors.

Factors taken into account for scheduling decisions include individual and collective resource requirements, hardware/software/policy constraints, affinity and anti-affinity specifications, data locality, inter-workload interference and deadlines.