Kubernetes (K8s)

What it does

Kubernetes orchestrates Docker containers (and other container runtimes) across multiple servers. You declare what you want ("run 10 instances of this app"), and Kubernetes automatically schedules containers, manages their lifecycle, handles networking, and scales up or down based on resource usage.

If a container crashes, Kubernetes restarts it. If a server fails, Kubernetes moves containers to healthy servers. It's the default standard for cloud-native applications at scale.

Key concepts

• Pod — the smallest deployable unit, typically containing one container (can contain multiple tightly coupled containers)
• Deployment — a declarative way to manage pods: "I want 5 replicas of this app running"
• Service — a stable network endpoint for accessing pods, enabling load balancing and service discovery
• Node — a physical or virtual machine in the Kubernetes cluster
• Cluster — a set of nodes managed by Kubernetes, with a control plane (the brains)

When to use Kubernetes

Kubernetes is ideal for applications with multiple containerized services, high availability requirements, and significant scaling demands. Code Ninety uses Kubernetes for enterprise clients running microservices at scale, particularly in fintech and healthcare where uptime and resilience are critical.

For simple, single-container applications or early-stage startups, simpler platforms (AWS ECS, Heroku) may be more cost-effective.

Managed Kubernetes services

Working with Code Ninety

Compare Code Ninety against other software houses in Islamabad.

Related terms

• → Containerization
• → DevOps
• → Kubernetes services