Serverless SurrealDB on Cloudflare scans 4.5B rows — 350GB — in one second
mattturck · x · 2026-11-04
With the launch of SurrealDB Earth — serverless SurrealDB on Cloudflare that starts on query arrival and scales to zero — its developer ran a real stress test: counting a 5-billion-row table.
- In a single second it read over 4.5 billion rows (90% of the table), more than 350 GB of data — all from a real run, not projections.
- No cluster provisioning or sizing needed; parallelism appears on demand and disappears after, running the full SurrealDB engine with ACID transactions.
- Not perfect yet: the slowest 0.4% of the work drags the full count out to several seconds, though the cause is understood.
Shared by Matt Turck, it's a striking datapoint on how far serverless databases can be pushed.
More from Infra
- Arm earnings: data center royalties double as selling own chips pits it against licensees — tengyanAI · 2026-11-05
- Open-source Local Office adds Snapdragon NPU support for local AI on Windows ARM64 — HowDevelop · 2026-11-05
- llama.cpp ships v0.9.0 as ggml team gears up for first major v1.0.0 release — ggerganov · 2026-11-05
- llama.cpp ships v0.9.0, first major v1.0.0 release weeks away — ggerganov · 2026-11-05
- llama.cpp named top open source project in GitHub Octoverse 2026 — ggerganov · 2026-11-05
- Google's free GPU masterclass breaks down Tensor Cores, caches and HBM for LLMs — mdancho84 · 2026-11-05