Advanced · 5h 2m · 20 lessons
Microservices
Split a big app into small, independent services that scale on their own.
What you'll learn
- ✦Explain how microservices differ from a monolith, and what the split gains and costs
- ✦Draw service boundaries around business capabilities, using bounded contexts
- ✦Choose a synchronous call (REST or gRPC) or asynchronous messaging for each interaction between services
- ✦Explain what an API gateway and service discovery each do, and spot when a system needs them
- ✦Design how services own and share data with database-per-service, sagas and eventual consistency
- ✦Apply timeouts, retries and circuit breakers, and judge when microservices are worth their cost
- ✦Plan how to package, deploy and observe many services with containers, Kubernetes and tracing
Part of these career paths
- Software Architecture path · course 3 of 3
Certification projects
Build 4 projects to earn your certificate
Hands-on work that proves you can apply what you learned — part of the certificate requirements.
Syllabus
Module 1 · 1h 3m
What are microservices? Free
What you'll be able to do
- Can tell a monolith from microservices by what deploys alone and who owns the data
- Can explain what teams gain by splitting, and work out what independent scaling saves
- Can calculate how a chain of network calls lowers availability, and name the running costs of a split
- Can draw first service boundaries around business capabilities
Module 2 · 1h
How services talk Locked
What you'll be able to do
- Can choose a synchronous call or an asynchronous message for each step of a flow
- Can pick a queue or a topic, and explain why consumers must cope with duplicates
- Can explain what an API gateway does, keep it thin, and spot when a BFF helps
- Can explain how a caller finds a healthy copy of a service whose address keeps changing
Module 3 · 44m
Data across services Locked
What you'll be able to do
- Can keep each service's data private and still show data from two services on one screen
- Can design a screen and an event flow that cope with eventual consistency, using an outbox
- Can plan a saga with compensating actions and a safe order of steps
Module 4 · 45m
Resilience Locked
What you'll be able to do
- Can set a timeout and a retry plan that fit inside a caller's time limit
- Can explain when a circuit breaker opens, what it does while open, and how it recovers
- Can make a handler idempotent so a duplicate request changes nothing
Module 5 · 43m
Operating them Locked
What you'll be able to do
- Can explain what containers and Kubernetes do, and size the copies a service needs at its peak
- Can use metrics, traces and logs to find where and why a request across services failed
- Can ship one service safely with backward-compatible changes and a canary rollout
Module 6 · 47m
Making the call Locked
What you'll be able to do
- Can decide whether a team should split into microservices, stay a modular monolith or take a middle path
- Can plan a strangler fig migration, starting with a low-risk service and its data
- Can choose a next topic to study, and a practice project, from a real problem