Mastering Redis: Core Data Structures and Practical Commands Explained

Redis is an open‑source, high‑performance key‑value store often described as a "data structure server" because it supports multiple mutable data structures accessed via a simple TCP client‑server protocol.

Six Core Data Structures

Redis provides six fundamental data types:

String : basic key‑value pairs, supporting binary data up to 512 MB. Commands: SET, GET, EXISTS, DEL, SETNX, GETSET, INCR, INCRBY.

List : a linked list (similar to Java LinkedList) offering O(1) push/pop at both ends. Commands: LPUSH, RPUSH, LPOP, RPOP, LRANGE, LINDEX.

Hash : a map of fields to values (akin to Java HashMap) implemented with an array of buckets and linked‑list chaining. Commands: HSET, HGET, HGETALL, HMSET, HDEL. Internally Redis maintains two hash tables to enable incremental rehashing.

Set : an unordered collection of unique elements (similar to Java HashSet). Commands: SADD, SMEMBERS, SISMEMBER, SCARD, SPOP.

Sorted Set (zset) : a set of unique members each associated with a double‑precision score, ordered by score. Implemented using a skip‑list for fast range queries. Commands: ZADD, ZRANGE, ZREVRANGE, ZRANK, ZSCORE, ZREM, ZRANGEBYSCORE.

Stream : introduced in Redis 5.0, a log‑structured data type designed for message‑queue use cases. Each entry has a unique timestamp‑sequence ID and stores field‑value pairs like a hash. Streams support consumer groups, pending entry lists (PEL), and both blocking and non‑blocking reads. Key commands: XADD, XGROUP CREATE, XRANGE, XREADGROUP, XACK, XDEL, XLEN.

String Operations

Setting and retrieving values is done with SET and GET. Overwrites replace previous values, while EXISTS and DEL test and delete keys. Expiration can be added via EXPIRE or combined commands like SETEX. Atomic increments are performed with INCR and INCRBY.

List Operations

Lists act as double‑ended queues. LPUSH / RPUSH insert at the head or tail; LPOP / RPOP remove elements. LRANGE retrieves a slice, and LINDEX accesses an element by index.

Hash Implementation Details

Redis hashes consist of two dictht tables. When the number of entries reaches the size of the first table, Redis allocates a new table (size doubled) and migrates entries incrementally to avoid long pauses. Rehashing proceeds during normal hash operations and a background task.

Set Operations

Sets store unique members. Adding duplicates returns 0. SMEMBERS lists all members (order is undefined), and SISMEMBER checks membership.

Sorted Set (zset) Details

Sorted sets guarantee uniqueness of members while ordering them by a floating‑point score. The underlying skip‑list provides O(log N) range queries. Example commands demonstrate adding members with scores, retrieving them in ascending or descending order, getting ranks, and removing members.

Stream Mechanics

Streams store messages in a linked list where each entry has a unique ID of the form milliseconds‑sequence. Consumers read entries via XREAD or XREADGROUP. Consumer groups maintain a last_delivered_id cursor and a pending entries list (PEL) to track unacknowledged messages. Commands such as XACK acknowledge processing, while XDEL removes entries without affecting the overall length.

Typical usage patterns include:

Message ID generation (auto‑generated or client‑specified).

Iterating over a range of IDs with XRANGE or XREVRANGE.

Blocking reads with BLOCK option for real‑time consumption.

Group consumption for parallel processing without duplication.

Monitoring pending messages and handling retries.

Conclusion

Understanding Redis's core data structures and their commands enables developers to choose the most appropriate type for caching, queuing, ranking, or real‑time data streaming, while leveraging Redis's in‑memory speed and atomic operations for reliable backend services.