Full-text search over all 49M HN items runs surprisingly fast on pure Neon Postgres
matei_zaharia · x · 2026-09-15
A developer built a full-text search over all of Hacker News (49M stories, comments, jobs & polls, synced hourly) that is surprisingly fast with no Next.js-level cache — all the magic happens at the Neon Postgres layer.
- Every keystroke is a live query against Neon Postgres, with BM25 ranking (powered by Lakebase Search) running inside the database.
- No connection pool, just SQL-over-HTTP round trips; pgprewarm loads the table and every index into cache.
- Partial BM25 indexes per type use predicates matching the WHERE clause; deployed on Vercel.
- The takeaway: "just Postgres doing Postgres things, close to the compute."
More from Infra
- Google Cloud and Inferact partner to make TPU a first-class vLLM target — vllm_project · 2026-09-15
- AI panic won't hurt chip sales: TL20 down 4% Monday but still up 53% this year — TiernanRayTech · 2026-09-15
- banteg's 11-byte EVM party trick: 16.78M gas projects to ~10TB of opcode trace — banteg · 2026-09-15
- Microsoft reportedly seeks ~38GW of data center capacity by 2032 — VraserX · 2026-09-15
- Memphis residents say noise from Musk data-center power plant is costing them sleep — LuizaJarovsky · 2026-09-15
- SSD prices are rising as AI demand breaks decades of falling storage costs — lemire · 2026-09-15