Presented by:

E4c4833d3a1ec1685839f37aea0caa28

Timofey Kulin

from YDB

Checkout the slides

YDB is a strongly consistent distributed database for mission-critical workloads with more than a decade of history that has been published to open-source in 2022. It was built from scratch with its own SQL dialect, API and query execution layer. We are happy to present the Postgres compatibility mode that has been added recently in response to open-source user feedback.

A few words about YDB architecture and internals:

  • Distributed storage layer
  • Tablet as a building block for scale, fault-tolerance and functionality
  • Maintaining consistency even if the whole datacenter goes down
  • Production scale examples (petabytes of data, millions of rps, thousands of servers in production cluster)

About PostgreSQL compatibility implementation:

  • Initial research
  • Good point for open-source projects
  • High-level implementation principles
  • Testing methodology
  • Performance optimizations for Postgres queries
  • Scaling read and write in postgres-like workloads, with detailed examples

Date:
2024 February 29 - 12:20
Duration:
40 min
Room:
Robusta + Arabica
Conference:
PGConf India, 2024
Language:
Track:
Database Engine Developers
Difficulty:
Medium

Happening at the same time:

  1. Unlocking the Power of Vector Similarity Search with pgvector
  2. Start Time:
    2024 February 29 12:20

    Room:
    Grand Victoria A

  3. Postgres load balancing is secretly broken: The cancellation problem
  4. Start Time:
    2024 February 29 12:20

    Room:
    Grand Victoria B