Resource requests and limits

As we were exploring the features of the K8s scheduler previously, have you wondered how Kubernetes knows what is the best node in the cluster for a particular pod? If we create a Deployment with no affinity settings, topology constraints, or node selectors, how can Kubernetes decide what is the best location in the cluster for the application we want to run?

By default, K8s is not aware of how many resources (CPU, memory, and other) each container in a scheduled pod requires to run. Therefore, for Kubernetes to make the best scheduling decisions, we need to make K8s aware of what each container requires for normal operation.

Resource requests

A resource request is an optional specification of how many resources each container in a pod needs. Containers can use more resources than requested if the node where the Pod runs has available resources. The specified request amounts will be reserved on the node where the pod is scheduled.

Kubernetes...