sys0Early access
Back to your course

Part 1 · Foundations

Part 2 · Distributed systems

  1. Overview6 lessons · 91 min
  2. 1Range vs hash partitioningRead
  3. 2Consistent hashingRead
  4. 3Hot keysRead
  5. 4Queries across shardsRead
  6. 5IDs without a central counterRead
  7. 6Lab: Reshard without downtimeLab

Part 3 · Under the hood

sys0Early access

Module 14 / Partitioning in depth / Overview

Module 14 6 lessons 91 min

Partitioning in depth

Sharding by city works until Mumbai has ten times the orders of any other city, and adding a shard means moving half the data.

In this module

5 Read · 1 Lab

  1. 14.1Range vs hash partitioningHow each method splits the data, and what it does to your queries.Read12 min
  2. 14.2Consistent hashingAdd a machine and move only a small part of the data.Read15 min
  3. 14.3Hot keysWhen one key gets most of the traffic, like a very popular restaurant.Read12 min
  4. 14.4Queries across shardsSecondary indexes, scatter-gather and request routing.Read12 min
  5. 14.5IDs without a central counterSnowflake IDs, UUIDs, and why order sometimes matters.Read10 min
  6. 14.6Lab: Reshard without downtimeAdd shards under live traffic without losing orders.Lab30 min
Start with 14.1 · Range vs hash partitioning
Then · Module 15Messaging: Kafka and pub/sub