Also known as: K8s

Kubernetes (often “K8s”) is an open-source system for automating the deployment, scaling, and management of containerized applications across a cluster of machines.¹ It was created at Google and released in 2014, and is now stewarded by the Cloud Native Computing Foundation.²

Overview

Kubernetes groups containers into pods, schedules them onto a pool of worker nodes, and continuously reconciles the cluster toward a desired state you declare. If a node dies, it reschedules the affected pods elsewhere; if load rises, it can add replicas. It also wires up service discovery and internal load balancing, rolling updates, and self-healing restarts. The containers themselves are usually built with Docker or another compatible runtime.

Where it fits

Kubernetes is the standard answer to running many containers reliably at scale, providing high availability and elastic scalability on top of plain virtualization. It is powerful but operationally heavy — overkill for a single service. A lone GopherTrunk capture node needs nothing like it, but a fleet of nodes feeding a shared back end could be coordinated by Kubernetes.

Sources

1. Kubernetes — Wikipedia, on Kubernetes’ design and history. ↩

2. Kubernetes — the project’s official site. ↩

See also

• Container
• Docker
• Load balancer
• High availability
• Scalability
• Virtualization