Redis Deep Dive: Why It’s Fast, Common Pitfalls, and Best Practices

Why Use Redis

Redis is chosen mainly for performance and concurrency, offering fast in‑memory operations and features like distributed locks.

Performance Benefits

Cache slow‑changing, expensive SQL results in Redis to serve subsequent requests instantly.

Concurrency Benefits

Under high load, routing requests through Redis prevents database connection overload.

Why Single‑Threaded Redis Is Fast

It relies on pure memory operations, a single thread avoiding context switches, and non‑blocking I/O multiplexing (select, epoll, kqueue).

An analogy of delivery workers illustrates the difference between the traditional thread‑per‑connection model and I/O multiplexing.

Redis Data Types

String : simple set/get, often for counters.

Hash : stores structured objects, useful for session‑like data.

List : implements simple queues and pagination.

Set : stores unique values and supports set operations.

Sorted Set : includes a score for ranking, useful for leaderboards, delayed tasks, and range queries.

Expiration and Memory Eviction

Redis uses a combination of periodic and lazy deletion. Periodic checks run every 100 ms on random keys; lazy deletion occurs when a key is accessed.

When memory is exhausted, eviction policies such as allkeys-lru, volatile-lru, noeviction, etc., determine which keys are removed.

# maxmemory-policy volatile-lru

Database‑Cache Double‑Write Consistency

To reduce inconsistency, update the database first, then delete the cache; if cache deletion fails, use a compensating mechanism like a message queue.

Cache Penetration and Avalanche

Penetration : use mutex locks, asynchronous updates, or Bloom filters to filter invalid keys.

Avalanche : add random jitter to TTLs, use mutex locks, or employ a double‑cache strategy with a warm‑up process.

Key Concurrency Issues

For unordered operations, a distributed lock is sufficient. For ordered updates, include timestamps or serialize operations with a single thread or queue.

Summary

The article summarizes common Redis questions encountered in practice and interviews, covering why to use Redis, its speed, data structures, expiration, eviction, consistency, and mitigation of cache‑related problems.