Analytics and AI with PostgreSQL - Mission Possible!
summary
Claims a framed product: EDB Postgres Lakehouse promises analytical queries up to 30× faster at 5× lower cost by separating compute from storage, keeping data in Parquet/Iceberg on object stores and using a Postgres front end. Demo shows PGAA/PGFS moving tables to S3 and a 6M‑row analytic query finishing in ~12s, replication via EDB Postgres Distributed, vector work via PGVector and a new AIDB extension for in‑db embeddings and SQL retrievers, plus a roadmap for GPU acceleration.
AI-written summary · about
chapters (10)
- 0:00PostgreSQL and AI analytics
- 3:20Traditional data warehousing challenges
- 4:03EDB Lakehouse for real-time analytics
- 10:08Live demo: analytics with PostgreSQL
- 12:38AI fundamentals with PostgreSQL
- 17:49Chatbot with vector embeddings
- 25:24AIDB extension for AI workflows
- 26:50Live demo: AIDB extension
- 30:50PostgreSQL analytics roadmap
- 32:20Q&A: performance and editions
related talks
34mAnalytics and AI with PostgreSQL - Mission Possible!
50mFrom Transactions to Intelligence: Postgres for Open AI & Analyti
27mBuilding a PostgreSQL data warehouse
49mCan PostgreSQL Compete with Analytics Specialists?
27mPutting AI in Production
19mFrom OLTP to OLAP: Is PostgreSQL Eating Analytics Too?
37mRound Table Discussion
21mAnalytics in Postgres –A decade in the making