QuillSQL
An educational relational database (RDBMS) written in Rust, designed for teaching database systems concepts inspired by CMU 15-445.
What is QuillSQL?
QuillSQL is an open-source relational database management system (RDBMS) written in Rust, designed as an educational tool for learning database system internals. It implements SQL query processing, transaction management, B+Tree indexing, buffer pooling, and ARIES-inspired recovery to solve the problem of understanding how production databases work under the hood.
Students, educators, and researchers studying database systems, as well as developers interested in database internals or building prototypes. It's particularly suited for those following curricula like CMU 15-445.
Developers choose QuillSQL for its clean, modular architecture that mirrors academic database courses, its implementation in modern Rust for safety and performance, and its focus on being a hackable codebase for experimentation without the complexity of industrial databases.
Overview
MLIR query compiler and JIT execution engine
Use Cases
Best For
- Learning database system architecture and implementation
- Teaching university-level database courses (e.g., CMU 15-445 style)
- Prototyping new database features or storage engines
- Experimenting with transaction isolation levels and locking
- Understanding write-ahead logging (WAL) and crash recovery
- Benchmarking and profiling database component performance
Not Ideal For
- Production systems requiring full SQL compliance, especially ALTER TABLE operations for schema evolution.
- Applications that depend on outer joins (LEFT/RIGHT/FULL) or complex arithmetic expressions in queries.
- Teams needing a database with extensive third-party integrations, ORM support, or management tools for rapid development.
- Projects where stability and minimal configuration are priorities over educational value and hackability.
Pros & Cons
Pros
The codebase has clear module boundaries and pluggable components like buffer pool and WAL, making it ideal for classroom assignments and research prototypes, as highlighted in the 'Teaching & Research Friendly' section.
Supports BEGIN/COMMIT/ROLLBACK, isolation levels, and row/table locking, providing a solid foundation for learning database concurrency and recovery mechanisms.
Uses io_uring-based asynchronous I/O and innovative buffer management with LRU-K and TinyLFU eviction, offering real-world insights into database optimization and hot-path performance.
Implements ARIES-inspired WAL and recovery with features like full-page writes and idempotent replays, detailed in the README, for deep dives into crash recovery and durability.
Cons
The README admits missing key features like ALTER statements, outer joins, and arithmetic expressions, limiting its use for complex or evolving database schemas.
As an educational tool, it lacks the driver libraries, ORM integrations, and community tooling found in established databases, making it unsuitable for seamless integration into existing stacks.
The project encourages 'experimental DB hacking' in its community section, indicating potential API changes and less stability compared to production-grade databases.
Frequently Asked Questions
Related Projects
SurrealDBA scalable, distributed, collaborative, document-graph database, for the realtime web
QdrantQdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
RustFS🚀2.3x faster than MinIO for 4KB object payloads. RustFS is an open-source, S3-compatible high-performance object storage system supporting migration and coexistence with other S3-compatible platforms such as MinIO and Ceph.
NeonNeon: Serverless Postgres. We separated storage and compute to offer autoscaling, code-like database branching, and scale to zero.
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.