PlanetScale Details Massively Parallel Backups: Restoring PB-Scale Databases at 50GB/s
bibryam · x · 2026-08-06
PlanetScale shared its massively parallel backup and restoration approach for petabyte-scale sharded databases (Postgres and MySQL). By spinning up temporary, backup-specific nodes for each shard, the system can pull data from object storage and replay Write-Ahead Logs (WAL) independently and in parallel.
This technique delivers massive speedups: restoring a 32TB dataset takes roughly 22 hours with a single stream, but only 42 minutes when parallelized across 32 shards. The architecture achieves backup rates exceeding 50GB/s without impacting production workloads.
More from Infra
- Cloudflare Emerging as the Agent Cloud via Rapid Iteration and Weird Bets — threepointone · 2026-08-06
- How Open-Source AI Inference Became Critical Infrastructure: a16z Podcast — a16z Podcast · 2026-08-06
- Calling for Decentralized AI: From the Cloud Back to a $5K Local Personalized AGI — Dan_Jeffries1 · 2026-08-06
- Maple 20B Hits 9,885 tokens/s on a Single NVIDIA GH200 in Concurrency Test — MaziyarPanahi · 2026-08-06
- Running 28.9M LLM on $8 ESP32: 10 tok/s Fully On-Device — FinanceYF5 · 2026-08-06
- Running Minimax Video on 12GB VRAM: 22-Min Render Tested — Svan_Derh · 2026-08-06