Docker with a focus on microservice-oriented architectures. The goal: connect both worlds — on one side, what defines a microservice and why it matters; on the other, how Docker and containers fit into that picture for building autonomous, isolated, and scalable systems.
What is a Microservice?
A microservice does one thing and does it very well. It’s a self-contained unit of functionality. In the talk I broke down the characteristics that typically define this architectural style:
- Autonomous — Identified by a unique location (URL). It’s a self-contained unit.
- Isolated — You can modify, test, and deploy it without impacting other areas of the system.
- Elastic — Scales independently: vertically (more resources) or horizontally (more instances).
- Resilient — Fault tolerant and highly available.
- Responsive — Responds in a reasonable amount of time.
- Message-oriented — Relies on asynchronous message-passing to establish boundaries between components.
- Programmable — Exposes APIs for developers, administrators, and applications composed from multiple services.
- Automated — Its lifecycle is managed through automation: dev, build, test, staging, production, and distribution.
Are Microservices the Silver Bullet?
No. They’re a tool, not a religion. The talk covered when they help and when they add complexity. The key is understanding the trade-offs: more services mean more network latency, more operational complexity, and challenges with distributed transactions. Not every project needs microservices from day one.
Docker Concepts
I moved on to Docker basics: containers, images, Docker Compose for orchestrating multiple services, and how they fit into a development and deployment flow. I also mentioned Docker Machine and Docker Swarm for more advanced setups.
Demos
I showed live demos during the talk, including a load balancer with Docker. The code is available in the repos I shared.
Resources
Demos and articles
- Docker Load Balancer Demo
- Divide and Conquer — Microservice Approach
- Ghost Docker
- Django Multiple Databases
Official documentation
- Docker
- Docker Compose
- Docker Machine
- Docker Swarm
- GitHub GraphQL API — useful for integrating with modern APIs from your containers
If you want a hands-on deep dive into Docker itself — from Dockerfiles and volumes to networks and Docker Compose — check out the Docker Introductory Workshop I conducted at Rocka Labs.
Let’s keep building.