Redis Caching Strategy

Designs effective Redis caching architectures for application performance optimization. Covers key naming conventions, TTL policies, eviction strategies (allkeys-lru, volatile-ttl), cache invalidation patterns (write-through, write-behind, cache-aside), data structure selection (strings, hashes, sorted sets, HyperLogLog), memory optimization, and Redis Cluster configuration for horizontal scaling.

Usage

Describe your application's data access patterns, the data you want to cache, read/write ratios, consistency requirements, and current performance bottlenecks. Specify your Redis deployment (single instance, Sentinel, Cluster) and memory budget. The skill designs a caching strategy with key schemas, TTLs, and invalidation logic.

Examples

• "Design a caching strategy for an e-commerce product catalog with 100k products and 10:1 read/write ratio"
• "Implement cache-aside pattern for user session data with automatic cache warming on miss"
• "Create a Redis-based leaderboard using sorted sets that updates in real-time with 1M entries"
• "Design a multi-level cache (L1 in-process, L2 Redis) for API responses with consistent invalidation"

Guidelines

• Use descriptive, hierarchical key names with colons: app:entity:id:field (e.g., myapp:user:123:profile)
• Set TTLs on all cache keys to prevent memory leaks — even long-lived caches should eventually expire
• Use cache-aside (lazy loading) as the default pattern: check cache → miss → fetch from DB → populate cache
• Choose the right data structure: hashes for objects, sorted sets for ranked data, sets for membership checks
• Use MGET/MSET for batch operations and pipelines for multiple commands to reduce round trips
• Configure maxmemory and maxmemory-policy (allkeys-lru for cache, volatile-lru for mixed workloads)
• Implement cache stampede protection using locking (SETNX) or probabilistic early expiration
• Monitor cache hit rate with INFO stats — aim for 95%+ hit rate for a well-tuned cache layer