Comprehensive Redis Interview Questions and Answers

Redis (Remote Dictionary Server) is an in‑memory key‑value database that persists data asynchronously to disk, offering extremely high read/write performance (over 100,000 ops/sec) and supporting rich data structures such as String, List, Set, Sorted Set, and Hash.

Its advantages over memcached include support for multiple data types, larger value size (up to 1 GB), built‑in persistence, and the ability to act as a lightweight message queue, tag system, or session cache.

Redis stores all data in RAM, making memory the primary resource; eviction policies (noeviction, allkeys‑lru, volatile‑lru, allkeys‑random, volatile‑random, volatile‑ttl) determine how data is reclaimed when memory limits are reached.

Clustering solutions include Codis, Twemproxy‑compatible proxy, and native Redis Cluster, which uses 16 384 hash slots to distribute keys across master nodes; the slot‑to‑node mapping can be configured or auto‑generated.

Persistence is provided by RDB snapshots (fork‑based) and AOF logs (append‑only), with mixed persistence in Redis 4.0+; replication follows a master‑slave model using binary‑log‑like command replay.

Distributed lock implementations range from simple SETNX with EXPIRE , to Lua‑scripted atomic operations in Redisson, and Zookeeper’s sequential‑node approach, each with trade‑offs in performance and failure handling.

Cache challenges such as penetration, thundering‑herd (cache stampede), and avalanche are mitigated by caching empty results, using Bloom filters, applying mutexes (e.g., SETNX lock), randomizing TTLs, and employing multi‑level caches.

When to choose Redis vs. Memcached: Redis is preferred for complex data structures, persistence, high availability, and larger values; Memcached excels at pure KV workloads with massive data volumes, lower latency due to its thread‑pool model, and simpler memory management.

Performance tips include avoiding persistence on master nodes, placing replicas in the same LAN, limiting slave count under heavy load, using linear replication topology, and monitoring memory usage to trigger appropriate eviction strategies.