คอนเทนเนอร์และการจัดการระบบด้วย Kubernetes

Microservices multiply the number of deployable units. Containers package each service with its dependencies into a portable, isolated image that runs the same everywhere.

An image is the immutable blueprint; a container is a running instance of it. You build an image once and run many identical containers from it.

Running a few containers by hand is easy. Running hundreds across many machines — restarting crashes, balancing load, rolling out updates — needs an orchestrator.

Kubernetes schedules containers onto a cluster of machines and keeps them running. Core objects:

• Pod — one or more containers that run together
• Deployment — declares desired replicas
• Service — stable network endpoint

You declare what you want; Kubernetes makes reality match. Ask for 3 replicas and it keeps 3 alive, restarting any that die.

If a node fails or a container crashes, Kubernetes reschedules the pod elsewhere automatically. The system converges back to the declared state without manual intervention.

Kubernetes uses probes to know container state:

• liveness — restart if it is hung
• readiness — only send traffic when ready

The Horizontal Pod Autoscaler adds or removes pods based on metrics like CPU, letting microservices scale out under load and back in when quiet.

Deploying a new version replaces pods gradually, keeping the service available throughout. If the new version misbehaves, Kubernetes can roll back to the previous one.

Many serverless and Functions-as-a-Service platforms run on top of Kubernetes. Understanding pods, autoscaling, and probes demystifies what happens beneath a deployed function.

Each container declares a CPU and memory request (what it needs to be scheduled) and a limit (its hard ceiling). Correct values let the scheduler pack nodes efficiently and stop one greedy pod from starving its neighbors.

You learned the container platform behind scalable services:

• Containers package services portably
• Kubernetes schedules and self-heals them
• Probes guide restarts and traffic
• Autoscaling and rolling updates keep services elastic and available