Design a Microservices Architecture

Design a microservices architecture for the following system:

System description: [what the application does]
Current state: [greenfield / monolith being split / existing microservices being redesigned]
Scale requirements: [expected users, requests per second, data volume]
Team size: [number of engineers — affects how many services are practical]
Tech stack preferences: [languages, cloud provider, container orchestration]
Key non-functional requirements: [high availability / low latency / strong consistency / eventual consistency / other]
Budget constraints: [startup / scale-up / enterprise]

Provide a complete architecture design:

SERVICE DECOMPOSITION:
- Recommended services list with responsibilities
- Boundaries between services (what each owns)
- Rationale for each split (why not monolith, why not finer-grained)
- Services that are strong candidates and services to leave as a monolith initially

COMMUNICATION PATTERNS:
- Synchronous vs asynchronous communication decisions per service pair
- API gateway design
- Event bus or message queue recommendations
- Service discovery approach

DATA MANAGEMENT:
- Database per service pattern applied
- Recommended database type per service and why
- Handling distributed transactions (saga pattern or alternatives)
- Data consistency strategy

INFRASTRUCTURE:
- Container orchestration recommendation
- Service mesh consideration
- CI/CD pipeline per service
- Monitoring and observability stack

SECURITY:
- Authentication and authorisation between services
- Secrets management
- Network security between services

FAILURE HANDLING:
- Circuit breaker pattern implementation
- Retry and timeout strategies
- Graceful degradation approach

MIGRATION PLAN (if splitting a monolith):
- Strangler fig pattern application
- Order of extraction (which service to extract first and why)
- Rollback strategy

TRADE-OFFS:
- What this architecture does well
- What it does poorly
- When to reconsider this approach