10 Powerful Redis Use Cases Beyond Simple Caching

Redis is often praised for caching, but its rich data structures enable many other backend solutions. Below are ten common scenarios, each with a brief description, relevant Redis commands, and example outputs.

1. Login Authentication

Store one‑time verification codes and tokens with SETEX to leverage key expiration.

redis> setex captchalogin|13436669876 60 3456
"OK"
redis> get captchalogin|13436669876
"3456"
redis> setex tokencheck|12345 86400 54321
"OK"
redis> get tokencheck|12345
"54321"

Use cases:

Send SMS code, store captchalogin|phone for 60 seconds.

Validate within the window; otherwise the code expires.

After login, store tokencheck|userId for 24 hours.

Subsequent requests verify the token.

2. Counters

High‑concurrency counters (likes, inventory) are efficiently handled with INCR / DECR and their BY variants.

redis> set article1 0
"OK"
redis> incr article1
(integer) 1
redis> decr article1
(integer) 0
redis> incrby article1 2
(integer) 2
redis> decrby article1 2
(integer) 0

3. Fan Following (Set)

Redis SET provides unordered, deduplicated collections with native set operations for common social‑graph queries.

redis> sadd Tony Mary
(integer) 1
redis> sadd Tony Lynn
(integer) 1
redis> smembers Tony
1) "Mary"
2) "Lynn"
redis> sinter Tony Tom
1) "Mary"
redis> sunion Tony Tom
1) "Mary"
2) "Lynn"
3) "Eric"
redis> sdiff Tony Tom
1) "Lynn"
redis> sdiff Tom Tony
1) "Eric"

4. Leaderboards (Sorted Set)

Sorted sets ( ZSET) store a score with each member, ideal for ranking and timelines.

redis> zadd 家电全品类 5.5 海尔
(integer) 1
redis> zadd 家电全品类 4.5 美的
(integer) 1
redis> zcard 家电全品类
(integer) 4
redis> zrevrange 家电全品类 0 -1 withscores
1) "海尔"
2) "5.5"
3) "美的"
4) "4.5"
...

5. Anti‑Scraping (Rate Limiting)

Limit actions per second using SET key "" EX 1 NX to allow only the first request.

redis> set createorder|userid|1234 "" EX 1 NX
"OK"   # first order succeeds
redis> set createorder|userid|1234 "" EX 1 NX
(nil)   # second order within a second is blocked
redis> set createorder|userid|1234 "" EX 1 NX
"OK"   # after a second, order succeeds again

6. Message Queue (List)

Use LPUSH for producers and RPOP for consumers to build a simple FIFO queue.

redis> lpush mybooks java
(integer) 1
redis> lpush mybooks mysql
(integer) 2
redis> rpop mybooks
"java"
redis> rpop mybooks
"mysql"

7. Browser History (Stack via List)

Push visited URLs onto a list and pop them to simulate the back button.

redis> lpush mybrowser sohu
(integer) 1
redis> lpush mybrowser sina
(integer) 2
redis> lpush mybrowser baidu
(integer) 3
redis> lpop mybrowser
"baidu"
redis> lpop mybrowser
"sina"
redis> lpop mybrowser
"sohu"

8. Distributed Lock

Acquire a lock with SET key value EX 10 NX. Release atomically with a Lua script that checks the owner.

redis> set mytasklock "tony" ex 10 nx
"OK"   # lock acquired by tony
redis> set mytasklock "tom" ex 10 nx
(nil)   # lock acquisition fails
redis> del mytasklock   # simple release (Lua needed for safety)

Production systems often use Redisson, which adds re‑entrancy and automatic renewal.

9. User Sign‑In (BitMap)

Store daily sign‑in status as bits; BITCOUNT yields total sign‑in days.

redis> setbit userid|1234|202312 0 1   # Dec 1 sign‑in
(redis) 0
redis> setbit userid|1234|202312 1 1   # Dec 2 sign‑in
(redis) 0
redis> bitcount userid|1234|202312
(integer) 3   # three days signed in

10. Website UV Statistics (HyperLogLog)

HyperLogLog provides approximate distinct counting with ~0.81 % error using only 12 KB.

redis> pfadd page1 user1
(integer) 1
redis> pfadd page1 user1
(integer) 0   # duplicate ignored
redis> pfcount page1
(integer) 4
redis> pfmerge page1and2 page1 page2
"OK"
redis> pfcount page1and2
(integer) 6

These patterns illustrate how Redis can solve a wide range of backend problems beyond simple caching, helping developers ace interview questions and build high‑performance services.