Databases & Data

Databases & Data covers the software that stores, indexes, and queries information, plus the tools that sit between your application code and that storage layer. This spans relational engines, key-value stores, columnar analytics databases, and the ORMs and query builders developers use to talk to them without writing raw SQL by hand. Projects here range from full database servers like TiDB and ClickHouse to embedded engines like DuckDB, in-memory stores like Redis, and platforms like Supabase that bundle a database with auth, storage, and APIs.

Choosing between these repositories depends on your access pattern and scale, not just the label on the tin. A high-throughput cache or session store calls for something like Redis, while analytical queries over large datasets favor a columnar engine such as ClickHouse or DuckDB. If you need horizontal scale with SQL compatibility, distributed databases like TiDB fit better than a single-node relational setup. For application development, an ORM like Prisma reduces boilerplate but adds an abstraction layer you need to trust for performance-sensitive queries.

When comparing projects, check:

• Data model: relational, document, key-value, graph, or columnar
• Consistency guarantees and how the project handles replication and failover
• Query language: SQL, a custom DSL, or an ORM API
• Deployment model: self-hosted binary, managed cloud service, or embeddable library
• Community activity, release cadence, and how breaking changes are handled

Most of these projects are open source under permissive or copyleft licenses, but managed hosting (like Supabase Cloud) often adds paid tiers on top of a free, self-hostable core. Read the license and check whether enterprise features are gated separately before committing to a stack.