Microservices Skill
Bonded to: architecture-patterns-agent (Secondary)
Quick Start
# Invoke microservices skill
"Decompose my monolith into microservices"
"Design API gateway for my services"
"Implement Saga pattern for distributed transactions"
Decomposition Strategies
| Strategy |
Best For |
Complexity |
| By Business Capability |
Clear domains |
Medium |
| By Subdomain (DDD) |
Complex domains |
High |
| By Team |
Conway's Law |
Medium |
| Strangler Fig |
Migration |
Low |
Service Decomposition Example
E-commerce Monolith → Microservices
├── user-service # Authentication, profiles
├── product-service # Catalog, inventory
├── order-service # Orders, checkout
├── payment-service # Transactions
├── notification-service # Email, push, SMS
└── api-gateway # Routing, auth, rate limiting
Communication Patterns
Synchronous (REST/gRPC)
# Service-to-service call
import httpx
async def get_user_from_user_service(user_id: str):
async with httpx.AsyncClient() as client:
response = await client.get(f"http://user-service/users/{user_id}")
return response.json()
Asynchronous (Events)
# Event publishing
from kafka import KafkaProducer
producer = KafkaProducer(bootstrap_servers=['kafka:9092'])
def publish_order_created(order):
producer.send('order-events', {
'type': 'ORDER_CREATED',
'order_id': order.id,
'user_id': order.user_id
})
Anti-Patterns to Avoid
| Anti-Pattern |
Sign |
Solution |
| Distributed Monolith |
Tight coupling |
Define bounded contexts |
| Shared Database |
Multiple services, one DB |
Database per service |
| Chatty Services |
Too many sync calls |
Use async messaging |
| Data Inconsistency |
No transaction strategy |
Implement Saga |
Troubleshooting
| Issue |
Cause |
Solution |
| Cascading failures |
No resilience |
Circuit breakers |
| Data inconsistency |
Distributed tx |
Saga pattern |
| High latency |
Chatty calls |
Batch requests, cache |
Resources
pluginagentmarketplace