Kubernetes

Kubernetes is a container orchestration platform. It is built to scale and is fault-tolerant. It can automatically scale containers when workloads are higher and scale down when workloads are lower. It can also automatically "fix" failed containers with a replica of the container. This is done by maintaining a set number of replicas of each container.

A system deployed on Kubernetes is called a cluster. The brain or the central Kubernetes deployment is called the Control Plane. It exposes several APIs that can handle requests to manage internal and external clusters. It contains a key-value pair database called etcd. This is used to manage the database. It manages a set of nodes (machines). Each node contains a Kubelet: a small application to communicate with the Control Plane. Each machine contains pods (containers).

As workload increases, it can scale horizontally by adding more nodes. Achieves High Availability by maintaining the replica set.