Art of Codeartofcode
Lesson 11 of 15

35. Design a Leaderboard (Top-K)

1
The interview prompt
2
35.1 What do you ask first?
3
35.2 How big is this, really?
4
35.3 What are the doors?
5
35.4 What do we store?
6
35.5 What does the whole system look like?
7
35.6 How do sorted sets work for this?
8
35.7 How do you handle real-time updates?
9
35.8 How do you handle ties?
10
35.9 How do you shard a leaderboard?
11
35.10 What would you say in the last five minutes?
12
Score yourself
13
Variations to practise
14
Sources
15
Code lab
35. Design a Leaderboard (Top-K)

35.10 What would you say in the last five minutes?

3 min read+10 XP

Step 6: Wrap-up (5 minutes)

  • Bottlenecks: write bursts from popular players (batching, sharding, pipelining) and refresh traffic (cached top 100, cached per-user rank until the next batch).
  • Single points of failure: Redis shards (replicas; rebuildable from teams and player points), the batch scheduler (run two with a lock or leader, Chapter 17).
  • Correctness for money: the final job recomputes from the verified event list, applies tie and prize rules, and writes an audited table; live ranks are informational.
  • Next with more time: friends' leaderboards (a small sorted set per group, or rank lookups for a list of friends), private leagues with instant updates, and historical leaderboards stored cheaply after the match.

Memory code: "Live is fast and roughly right; final is slow and exactly right."