Redis Overview: Features, Use Cases, Data Types, Persistence, Distributed Locks, and Performance

Redis is an open‑source, ANSI‑C written, network‑enabled, in‑memory (with optional persistence) key‑value database that provides APIs for many languages.

Typical use cases include high‑concurrency read/write, massive data volume, and scenarios demanding high scalability.

Key functionalities are caching, distributed locking, data persistence, transactions, and message queuing.

Compared with Memcached, Redis supports richer data types, is faster, and can persist data.

Redis runs single‑threaded because CPU is not the bottleneck; memory or network bandwidth are, making a single thread simple and efficient.

Cache penetration occurs when repeatedly querying non‑existent data; a common mitigation is to cache empty results with a short TTL (e.g., up to five minutes).

Supported data types are string, list, hash, set, and zset.

Popular Java clients include Redisson, Jedis, and Lettuce, with Redisson being officially recommended.

Jedis offers comprehensive command coverage, while Redisson provides distributed data structures and abstracts Redis usage, but lacks support for raw string operations, sorting, transactions, pipelines, and sharding.

To keep cache and database consistent, set appropriate TTLs and synchronize updates, optionally using transactions.

Redis persistence has two strategies: RDB snapshots at intervals and AOF which appends every write command.

Distributed locks are implemented using SETNX (set if not exists) to acquire a lock and DEL to release it; however, they cannot automatically handle lock timeout issues.

Memory optimization recommends using hashes to store related fields together, reducing the number of keys.

Eviction policies include volatile‑lru, volatile‑ttl, volatile‑random, allkeys‑lru, allkeys‑random, and no‑eviction.

Common performance problems are blocking during large RDB snapshots and suboptimal master‑slave replication placement; keeping replicas in the same LAN mitigates the latter.