How to Configure Big Data Management on Kubernetes

Back Next

Components in the Kubernetes Architecture

The following components are integral to the Kubernetes architecture:

Pods: A group of one or more containers with shared resources and instructions on how to run containers. Pods can communicate within the cluster using the service IP.; The following image shows the different pods and the various states in which a pod can exist:

Pod 1 contains a containerized application. Pod 2 contains a containerized application and the volume. Pod 3 contains a containerized application and two volumes. Pod 4 contains two volumes and three containerized applications.

Services: An abstraction that defines a logical set of pods and a policy by which to access the pods. A service in Kubernetes is a rest object that is similar to the pod. Kubernetes services support TCP and UDP for protocols, where the default is TCP. Pods are ephemeral but the services are more stable and you can communicate between different services. It is the Kubernetes service that routes traffic to the specific pod with the same application label.; The following image displays a pod with a service targeting the pod using the application label:; The image shows how a pod with a service A targets the pod with the application label as app=A.
Volumes: A directory that contains data that is accessible to the containers in the pod. A volume lasts longer any containers that run within the pod and data is preserved across container restarts. It also helps maintain the state within the cluster.
Namespaces: Separates workloads from each other along with the resource constraints on each. For example, the pod or service can have multiple developers with a namespace for each developer. To provide a similar workload for each developer, you can configure each namespace state to consume no more than 2 GB RAM.