Module 1 - Data Storage

Redis vs Memcached

Two popular in-memory caching solutions. Which one should you choose?

1Quick Comparison

Simple Analogy

Memcached: A simple, blazing-fast key-value store. Like a hash table in memory. Does one thing extremely well.

Redis: A Swiss Army knife. Key-value plus lists, sets, sorted sets, pub/sub, persistence. More features, slightly more overhead.

2Feature Comparison

Feature	Redis	Memcached
Data Types	Strings, lists, sets, hashes, streams	Strings only
Persistence	RDB, AOF	None (pure cache)
Replication	Built-in	None
Pub/Sub	Yes	No
Clustering	Redis Cluster	Client-side sharding
Multi-threading	Single (mostly)	Multi-threaded
Memory Efficiency	Good	Better for simple strings

3When to Use Each

Choose Redis

✓ Need data structures (lists, sets, sorted sets)
✓ Need persistence/durability
✓ Need pub/sub messaging
✓ Need replication/HA
✓ Rate limiting, leaderboards, sessions

Choose Memcached

✓ Simple key-value caching only
✓ Need multi-threaded performance
✓ Maximum memory efficiency for strings
✓ Already have Memcached expertise
✓ Pure cache, no persistence needed

4Performance Considerations

Memcached: Multi-threaded

Can use multiple CPU cores. Better for high-throughput simple caching on multi-core machines.

Redis: Single-threaded (mostly)

One main thread for commands. Still incredibly fast (100K+ ops/sec). I/O threads added in Redis 6.

Memory Overhead

Redis has more overhead per key due to data structure flexibility. Memcached is leaner for pure strings.

5Key Takeaways

1Redis is feature-rich: data structures, persistence, pub/sub, replication

2Memcached is simpler: pure cache, multi-threaded, memory efficient

3For simple caching only, Memcached may be slightly faster

4For anything beyond simple caching, Redis is the better choice

5Most projects choose Redis for its versatility and ecosystem

?Quiz

1. You need to cache database queries with automatic expiry. Which works?

2. You need a real-time leaderboard with rankings. Which should you use?