Partitioning, Sharding & Zero-Downtime Migration
Split data without splitting your team: shard keys, hot shards, cross-shard queries, resharding with logical shards, and moving a live table with no downtime.
About 95 minutes · 15 exam questions · 9 interview questions · 2 code labs
What you will learn
- Why split data at all — and why sharding comes last
- Native table partitioning in PostgreSQL (the step before sharding)
- Sharding strategies: hash, range, directory, geo
- Choosing the shard key — the decision you cannot cheaply undo
- Hot shards and the celebrity problem
- Cross-shard queries, joins and global secondary indexes
- Resharding: consistent hashing, logical shards, Vitess and Citus
- Federation (functional partitioning) and cells
- Zero-downtime data migration: dual writes vs CDC, shadow reads, cutover
- Generating IDs across shards: Snowflake and friends
Previous: Replication, Read Replicas & Replication Lag · Next: Database Firefighting: Six Production Scenarios