Managing resources in Kubernetes
Kubernetes provides the ability to proactively allocate and limit resources available to Kubernetes objects. In this section, we will discuss resource requests and limits, which form the basis for resource management in Kubernetes. Next, we explore namespace resource quotas and limit ranges. Using these two feature, clusters, administrators can cap the compute and storage limits available to different Kubernetes objects.
Resource requests and limits
kube-scheduler, as we discussed in Chapter 1, Kubernetes Architecture, is the default scheduler and runs on the master node. kube-scheduler finds the most optimal node for the unscheduled pods to run on. It does that by filtering the nodes based on the storage and compute resources requested for the pod. If the scheduler is not able to find a node for the pod, the pod will remain in a pending state. Additionally, if all the resources of the node are being utilized by the pods, kubelet on the node...
