Mvcc

An experimental MVCC implementation for SQLite that gives you row-level snapshot isolation instead of the usual page-level approach. Each transaction sees a consistent snapshot from its start time, with multiple row versions tracked in memory using lock-free skip lists. The big win is finer-grained concurrency, but it's explicitly not production-ready. Checkpoints still block everything including reads, there's no garbage collection so old versions pile up, and nothing survives a restart since log recovery isn't implemented. Useful if you're exploring concurrency control patterns or need to prototype snapshot isolation behavior, but treat it as a learning tool until those gaps get filled.

skills.sh installs: 1.2K
GitHub stars: 25.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add tursodatabase/turso --skill mvcc --agent claude-code

Installs into .claude/skills of the current project.

Files

MVCC Guide (Experimental)

Multi-Version Concurrency Control. Work in progress, not production-ready.

CRITICAL: Ignore MVCC when debugging unless the bug is MVCC-specific.

Enabling MVCC

PRAGMA journal_mode = 'mvcc';

Runtime configuration, not a compile-time feature flag. Per-database setting.

How It Works

Standard WAL: single version per page, readers see snapshot at read mark time.

MVCC: multiple row versions, snapshot isolation. Each transaction sees consistent snapshot at begin time.

Key Differences from WAL

Aspect	WAL	MVCC
Write granularity	Every commit writes full pages	Affected rows only
Readers/Writers	Don't block each other	Don't block each other
Persistence	`.db-wal`	`.db-log` (logical log)
Isolation	Snapshot (page-level)	Snapshot (row-level)