The open dataset behind whichdb.com: one JSON profile for every database we can find (active, dead and research systems), with its data model, workloads, architecture, license, deployment options, features, strengths, trade-offs and alternatives.
This repository holds data only. The website's code is not open source.
databases/<slug>.json one profile per database (844 today)
logos/<slug>.webp full-colour logos for databases without a Simple Icons mark
logos/sources.json where each logo came from
schema/profile.schema.json JSON Schema (draft 2020-12) every profile must satisfy
databases/postgresql.json is the reference example.
Clone the repo, or read files directly:
https://raw.githubusercontent.com/onecompiler/whichdb/main/databases/postgresql.json
The site also publishes a compact catalog at https://whichdb.com/data/index.json and each full
profile at https://whichdb.com/data/db/<slug>.json, with rankings mixed in.
Popularity metrics (the WhichDB Score) are not part of this dataset.
The profiles and the schema are licensed under CC BY 4.0. You may copy, adapt and use them for any purpose, including commercially, as long as you give credit, for example:
Data from WhichDB, licensed under CC BY 4.0.
Logos are not covered by that license. Every logo and brand mark is a trademark of its owner and is included only to identify the database. See NOTICE.
Spotted something wrong or out of date? Every page on whichdb.com has a suggest an edit link that opens the profile here. Fix it and open a pull request. Read CONTRIBUTING.md first; it explains the fields, how to research a change and how to validate it:
npm install
npm run validateMerged changes reach whichdb.com within a day.