Leap Ahead with Redis 6.2




Brian Sam-Bodden as Developer Advocate, Redis Labs

DaShaun Carter as Partner Solution Architect, Redis






English 🇺🇸


Beginner-Friendly Spring



  • ✅ Enthusiastic talk covering a wide range of Redis topics with practical examples with some ease.

  • ⛔ Some commands could be omitted in favor of discussing where Redis aims now (as said at the end of the session).

"The 'Most Loved' database in StackOverflow’s Developer survey for the 5th year in a row."

Redis (162 clients in 50 languages) stores data in memory, not on disk, which brings <1ms latency.

Spring Data is a family of projects giving a Java/Spring idiomatic way to access data from low-level constructs to high-level OO abstractions in either non/reactive or functional ways.

Spring Data Redis provides easy configuration and access to Redis through low-level connectivity via Lettuce & Jedis libraries, provides RedisTemplate as a high-level abstraction for Redis operations (Ops), and implements key-value mappings and repositories.


String: ValueOperations

ValueOperations can be performed through StringRedisTemplate#opsForValue and the maximum size of a Redis Key is 512MB and the Redis value is 512MB

  • redis-cli set stringKey stringValue (SET) sets a key-value.

  • redis-cli get stringKey (GET) gets a value by key.

  • redis-cli getset stringKey stringUpdatedValue (GETSET) gets a value by key and sets the value immediately in one operation.

  • redis-cli getdel stringKey (GETDEL) gets a value by key and deletes the key immediately.

  • redis-cli getex stringKey ex 3 (GETEX) gets and expires a key in a certain number of seconds.

  • redis-cli set stringKey stringValue exat 1662163200 (PXAT) sets and expires a key at the sprcifiec Unix time in seconds.

  • redis-cli set stringKey stringValue pxat 1662163200000 (PXAT) sets and expires a key at the specific Unix time in milliseconds.

  • redis-cli ttl stringKey (TTL) returns the remaining time to live of a key that has a timeout.

Hash: HashOperations

HashOperations (HSET/HMSET/HGETALL) can be performed through StringRedisTemplate#opsForHash, the hash maps identified by keys between string fields and string values closely resembles Java Map and can store over 4 billion field-value pairs.

  • redis-cli hset hashKey currentTime "$" (HMET) sets a field in the hash stored at the key to value.

  • redis-cli hmset hashKey status "Good" name "Redis" greeting "Hi" (HSMET) sets multiple fields in the hash but is deprecated as of Redis 4.0.0 in favor of HSET

  • redis-cli hget hashKey greeting (HGET) returns the value associated with field in the hash stored at key.

  • redis-cli hgetall hashKey (HGETALL) returns all fields and values of the hash stored at key.

  • redis-cli hrandfield hashKey 4 WITHVALUES (HRANDFIELD) returns an array of random distinct fields (if positive count) or random fields with possible duplicates (if negative count).

List: ListOperations

ListOperations can be performed through ListOperations<K, V> and they are implemented in Redis as a linked list but has enough commands to turn it into a stack, queue, or any linear storage and store over 4 billion entries.

  • Adding: Pushing in LPUSH/LPUSHX/RPUSH/RPUSHX, inserting before/after LINSERT, and setting the value at an index LSET.

  • Removing: Popping off LPOP/RPOP/BLPOP/BRPOP (including blocking operations), by value LREM and by index range LTRIM.

  • Accessing: By index LINDEX and by range LRANGE.

  • Between sits: Last from one/to fill in another RPOPLPUSH/BRPOPLPUSH and pop and then push LMOVE/BLMOVE (including blocking operations).

Set: SetOperations

SetOperations can be performed through SetOperations<K, V> and they are a collection of unique and unsorted string elements supported by operations such as union, intersection, and subtraction, most operations perform in constant time (O(n)).

Use cases: unique item management, tracking OPs, content filtering.

  • Adding SADD and removing SPOP/SREM.

  • Accessing SMEMBERS/SRANDMEMBERS and retrieving `SSCAN.


Sorted Set: ZSetOperations

ZSetOperations operations can be performed through ZSetOperations<K, V>, they are weighted sets, tuples with a value and a score, and elements are always taken by their score or in ranges, in spring Data Redis uses Set<TypedTuple<E>> data type.

Use cases: Priority queues, low-latency leaderboards, or secondary indexing in general

  • redis-cli zadd game1 100 "Frank" 740 "Jennifer" 200 "Pieter" 512 "Dave" 690 "Ana" (ZADD)

  • redis-cli zrange game1 0 -1 withscores (ZRANGE)

  • redis-cli zinter 2 game1 game2 withscores / redis-cli zinter 2 game1 game2 withscores aggregate max (ZINTER)

  • redis-cli zdiff 2 game1 game2 withscores

  • redis-cli zadd game1 100 "Foo" (ZADD)

Geo: GeoOperations

GeoOperations can be performed through StringRedisTemplate#opsForGeo, it is a sorted set as a latitude and longitude encoded into the score of the sorted set using the geo-hash algorithm.

Since it is still a sorted set, it is possible to use Z* commands

  • redis-cli GEOAD running-poi -94.238226 39.029377 "Lake"

  • redis-cli GEODIST running-poi "Lake" "Park"


Streams before Redis 6.2 could be only trimmed to an exact or approximate number of entries which was odd as it defies stream processing

The rule of thumb is that each entry in a stream must have a unique ID greater than any previously seen in the stream (Redis by default uses milliseconds timestamps for this) and now allows you to trim based on ID.

Future of Redis

Redis team is up to extending and complementing Spring Data Redis with:

  • Access to module commands via Spring’s Templates, multi-model object-mapping support, JSON object-mapping, and RediSearch integration, Redis Graph oriented-mapping, RediSearch integration for existing Redis Hash mapped entities.

  • Redis Modules Templates follow Spring Data Redis opsForXXX() pattern and provide a Native way to interact at the command level with RedisSON, RedisGraph, RediSearch, RedisAI, RedisBloom, and RedisTimeSeries