Scry

ScryData is a database change validation tool that captures real production query workloads and replays them against proposed database changes on a pinned copy of production data, providing reproducible proof of performance impact. It solves the problem of unpredictable database changes by predicting and proving query performance before deployment. The tool targets developers, data teams, and operations teams in companies of various sizes, especially those using PostgreSQL. It is positioned as an open-stack alternative to Oracle Real Application Testing, offering a transparent, reproducible approach with a free tier.

Key features

• Planner what-if analysis estimates query cost changes
• Replays real captured workload with real timing and concurrency
• Captures production queries transparently with bind parameters
• Pins data snapshot, stats, workload window, and Postgres version
• Provides reproducible receipts with regression and improvement details
• Supports schema migrations, version upgrades, index changes, and more
• Low proxy overhead (~100μs) safe for production traffic
• Zero-copy query parsing built in Rust