Caching Infrastructure

Caching Infrastructure

Core Idea: Redis provides fast in-memory caching for query results, service contracts, and session data with automatic TTL management.

Overview

The platform uses Redis 7 for caching:

• Query results (CQRS queries)
• Service contracts (service discovery)
• Session data (API Gateway)
• Read model materialization

Configuration

Development

redis:
  image: redis:7-alpine
  command: redis-server --appendonly yes --requirepass redis123
  ports:
    - "56379:6379"
  volumes:
    - redis-data:/data

Connection

REDIS_URL=redis://:redis123@redis:6379

Usage Patterns

Query Caching

@QueryHandler(GetUserQuery, {
  cache: {
    ttl: 60,  // Cache for 60 seconds
    key: (query) => `user:${query.userId}`
  }
})
export class GetUserHandler {
  async handle(query: GetUserQuery) {
    // Result automatically cached
    return await this.userRepo.findById(query.userId);
  }
}

Contract Caching

// Service discovery caches contracts
await redis.set(
  `contracts:${serviceName}:${version}`,
  JSON.stringify(contracts),
  'EX',
  3600  // 1 hour TTL
);

Best Practices

1. Set Appropriate TTLs

   • Frequently changing data: 10-60 seconds
   • Stable data: 5-60 minutes
   • Configuration: Hours to days

2. Use Cache Keys Wisely

   • Include relevant query parameters
   • Use namespacing: user:123, not just 123

3. Monitor Hit Rate

   • Target >80% hit rate for cached queries
   • Low hit rate indicates wrong TTL or cache keys

Monitoring

# Redis stats
redis-cli -a redis123 INFO stats

# Key space
redis-cli -a redis123 DBSIZE

# Memory usage
redis-cli -a redis123 INFO memory

Further Reading

• CQRS Pattern
• Redis Documentation