Skip to main content

Redis Sorted Sets

1. Why It Is Used​

  1. Redis sorted sets are a data structure that store unique elements, each associated with a floating-point score. The elements are automatically sorted based on their scores, making sorted sets useful for scenarios requiring ordered data retrieval and efficient range queries.
  2. If two members have the same score, Redis sorts them based on members’ lexicographical order.

2. Use Cases​

Sorted sets are commonly used for:

  • Leaderboards: Maintaining dynamically updated rankings (e.g., gaming scores, competition rankings).
  • Rate Limiting: Implementing sliding-window rate limiting to control API request rates.
  • Priority Queues: Managing tasks based on priority.
  • Ranking Systems: Keeping track of top performers or trending items in real-time applications.

3. Practice Examples​

Adding Elements to a Sorted Set​

ZADD <name_of_sorted_set> <score> <member>
ZADD leaderboard 100 "Alice"
ZADD leaderboard 200 "Bob"

Change value of score for a member​

ZINCRBY <name_of_sorted_set> <increment> <member>
ZINCRBY players 10 "Alice"
ZINCRBY players -10 "Bob"

Retrieving the Top Players​

returns members in reversed-rank order, with scores ordered from high to low. You have to specify starting and ending rank index positions with a start and stop parameters.

ZREVRANGE <key> <start> <stop> [WITHSCORES]
// for top 10 players
ZREVRANGE leaderboard 0 9 WITHSCORES

Retrieve the rank and the score of an individual player​

[ZRANK|ZREVRANK] <key> <member>

Removing a Player​

ZREM <key> <member> [<member> …]
ZREM leaderboard "Alice"

Union multiple set to create a global leaderboard​

ZUNIONSTORE destination numkeys key [key ...] [WEIGHTS weight
  [weight ...]] [AGGREGATE <SUM | MIN | MAX>]

It makes the union of multiple sorted sets passed as key and stores the result in location destination. Using WEIGHT option is to specify a multiplication factor for each set score before union, by default it is 1. Agreegate option is used to specify what we need from union, sum, max, min. If destination already exist will be overwritten

ZUNIONSTORE global_leaderboard 2 leader_board_game1 leader_board_game2 WEIGHTS 2 3

4. Time Complexity​

  • ZADD: O(log N) (Adding an element)
  • ZRANGE: O(log N + M) (Retrieving elements in a range, where M is the number of results)
  • ZRANGEBYSCORE: O(log N + M)
  • ZREM: O(log N) (Removing an element)
  • ZREVRANK: O(log N) (Getting reverse rank of an element)
  • ZUNIONSTORE: O(N) + O(M log M) (with N being the sum of the sizes of the input sorted sets, and M being the number of elements in the resulting sorted set.)

Redis sorted sets efficiently handle ordered data, making them ideal for real-time applications requiring fast insertions, deletions, and queries.