Breakout Session
summary
Postgres positions itself as a central data broker by shipping concrete features: JSONB 9.4 with benchmarks showing much smaller indexes and dramatically faster queries for typical workloads, pluggable extensions via create extension (PL/Python, PostGIS), new storage/index types like BRIN indexes for huge append‑only tables, and foreign data wrappers that join across Oracle, MongoDB, S3, etc. Work on auto‑sharding and parallel sequential scan 9.6 is underway; replication + FDWs enable cross‑store analytics today.
AI-written summary · about
chapters (24)
- 0:00Intro
- 3:45Outline
- 6:24Object-Relational (Extensibility)
- 10:24Example: ISBN Data Type
- 13:36The System Catalog Entry for INTEGER
- 14:36Not Just Data Types, Languages
- 16:07Available Languages
- 17:12Specialized Indexing Methods
- 17:45Index Types Are Defined in the System Catalogs Too
- 18:00Operators Have Similar Flexibility
- 19:50Other Extensibility
- 22:08Offshoots of Postgres
- 22:44Plug-In Is Not a Bad Word
- 26:35When Should Relational Storage Be Used?
- 26:57The Best of Both Worlds: Postgres
- 31:48Easy Relational Schema Changes
- 32:28Data Analytics
- 39:03Read-Only Slaves for Analytics
- 42:31Foreign Data Wrappers to Non-Traditional Data Sources
- 43:20Foreign Data Wrapper Example
- 44:174. Foreign Data Wrappers (Database Federation)
- 47:26Read and Read/Write Data Sources
- 50:36Postgres Centrality
- 50:54Postgres's Central Role
