Getting Started with Kubernetes: Reduce Dev Overhead and Optimize Resources

Preface

If you want to focus on development and reduce the mental load of low‑level resource configuration and deployment, or you want to make better use of underlying resources and monitor applications more conveniently, you should learn about Kubernetes. Kubernetes (also called k8s) is an open‑source system for automating deployment, scaling, and management of containerized applications.

Do I Need Kubernetes?

Kubernetes is a container orchestration system that handles:

Deploying and scheduling containers

Automatically managing application lifecycles

Balancing container and cluster resources

Configuring service discovery and traffic management

…

When an application consists of multiple services running in different containers, using Kubernetes is a good choice. For a monolithic application serving a stable user base, Kubernetes may not be necessary.

If you have more than one node, Kubernetes abstracts the underlying infrastructure and automatically migrates workloads based on node status, making a set of nodes appear as a single large node and simplifying development and deployment.

How to Use Kubernetes?

Environment Selection

minikube – suitable for learning k8s or local development and testing; it creates a Kubernetes environment on your own machine.

Kubernetes – appropriate for production deployments that require managing multiple nodes and clusters.

Resource Management

Kubernetes resources are objects in the cluster that represent different components and services of an application. Three common ways to manage them are:

kubectl – the official command‑line tool for interacting with the cluster; it offers many functions and is suited for users with Kubernetes experience.

YAML files – defining resources in declarative YAML improves readability and version control; a VS Code YAML plugin can provide templates for Kubernetes objects.

Graphical tools – tools such as Kuboard or Rancher provide visual interfaces that simplify configuration.

How Does Kubernetes Work?

Kubernetes architecture consists of a control plane and worker nodes.

Control Plane

etcd – a fast, distributed, consistent key‑value store that holds the desired state of resources.

API server – the sole component that talks directly to etcd; it exposes a RESTful CRUD interface. The API server does not modify resources itself; it records state changes in etcd and notifies clients (e.g., scheduler, controllers) to act.

Scheduler – watches for newly created pods via the API server and assigns each pod to an appropriate node.

Controller manager – runs various controllers (e.g., Deployment, Namespace) that ensure the actual cluster state converges toward the desired state stored in etcd. For example, the Namespace controller deletes all resources inside a namespace when the namespace is removed.

Worker Nodes

kubelet – manages and monitors containers on the node and continuously reports their status to the API server.

Container runtime – pulls images and runs containers.

kube-proxy – ensures that clients can reach services defined in the cluster via the Kubernetes API.

Recommended Resources

Kubernetes official documentation

Kubernetes in Action (Chinese edition)