📦

Microservices Architecture with Docker and Kubernetes

Apr 18, 2025

Lecture Notes on Microservices and Kubernetes

Introduction to Microservices

  • Microservices: Architectural style to develop a single application as a suite of small services
  • Each service runs its own process and communicates via HTTP or messaging
  • Developed and deployed independently

Advantages of Microservices

  • Scalability: Services can be scaled independently
  • Resilience: Failure of one service doesn't affect others
  • Flexibility: Different technologies can be used for different services
  • Faster Deployment: Smaller codebase to manage and deploy

Disadvantages of Microservices

  • Complexity: Difficult to manage due to decentralization
  • Network Latency: Communication over the network can cause delays
  • Data Management: Ensuring consistency across services is a challenge

Docker Overview

  • Docker: Platform to develop, ship, and run applications in containers
  • Containers: Lightweight, portable, and self-sufficient units of software
  • Dockerfile: Script containing instructions to build Docker images
  • Docker Hub: Cloud-based registry service for sharing Docker images

Kubernetes Overview

  • Kubernetes (k8s): System for automating deployment, scaling, and management of containerized applications
  • Cluster: Group of nodes where Kubernetes manages containers
  • Pods: Smallest deployable units in Kubernetes, which can contain one or more containers
  • Services: Abstractions to expose an application running on a set of Pods
  • Ingress: Manages external access to services, typically HTTP

Developing a Microservices Application

  • Platform Service: REST-based API to manage platforms
    • Uses SQL Server for database
    • Deployed as a Docker container
  • Command Service: REST-based API to manage commands
    • Uses in-memory database
    • Deployed as a Docker container
  • RabbitMQ: Message broker for asynchronous communication
  • API Gateway: Manages external access to services

Implementing Synchronous and Asynchronous Communication

  • Synchronous Messaging: Direct request/response communication (e.g., HTTP, gRPC)
  • Asynchronous Messaging: Event-driven communication using message brokers (e.g., RabbitMQ)

gRPC

  • gRPC: High-performance RPC framework using HTTP/2
  • Uses Protocol Buffers to define service contracts
  • Ideal for service-to-service communication

Kubernetes Deployment

  • Deployment Files: YAML files defining desired state of Pods, Services, etc.
  • kubectl: Command-line tool to interact with Kubernetes cluster
  • Persistent Volume Claims: Request storage resources in Kubernetes

Final Thoughts

  • HTTPS/TLS within Kubernetes clusters needs further exploration
  • Event processing could be more elegantly designed
  • Service discovery could simplify service intercommunication

This course covered the fundamentals of setting up a microservices architecture using Docker and Kubernetes and implementing both synchronous and asynchronous communication between services. The complexity and considerations involved in deploying microservices at scale were emphasized.

Full transcript