How PG Sharding Works in Yandex Taxi’s Processing
summary
Describes Yandex.Taxi’s custom PostgreSQL sharding solution inside its order processing core, including how data is organized to meet high load, strict availability, and low-latency requirements. Explains the practical approach used to handle load and implement a simple, reliable resharding mechanism.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 7:51Scaling problem
- 9:59PostgreSQL sharding
- 16:03Sharding-friendly principles
- 23:11Queries and resharding
- 28:32Pros and cons
- 31:19Summary
- 32:26Q&A
related talks
49mSharding: from scratch to Yandex Disk
39mMailing Adventures with PostgreSQL: Taming 650+ Shards and Surviving
52mTaking Storage for a Ride
45mThe Future of Postgres Sharding
51mYandex Database: Distributed Queries in the Clouds
42mSharding Postgres without doing it yourself
45mThe Future of Postgres Sharding
31mPostgreSQL Distributed & Secure Database Ecosystem Building