How Dolt Stores Table Data | DoltHub Blog
dolthub.com · 1,893 words · saved by 1 readers
Blog for DoltHub, a website hosting databases made with Dolt, an open-source version-controlled SQL database with Git-like semantics.
Dolt is Git for data . It’s a SQL database that lets you clone, branch, diff, merge, and fork your data just like you can with a filesystem tree in Git. This blog post explores one of the fundamental datastructures that underlies Dolt’s implementation of SQL tables. What We Need # Dolt currently supports a subset of SQL where every table must have a primary key. The SQL layer is implemented on top of a table storage layer that models a key-value store, where keys and values are both byte arrays. To efficiently implement the Git for data portions of Dolt , we need certain unique properties from
related reading
- Prolly Trees | DoltHub Blogdolthub.com
- Merklizing the key/value store for fun and profit | Joel Gustafsonjoelgustafson.com
- Git at any scalecursor.com
- Building a BFT JSON CRDTjzhao.xyz
- Build Your Own Databasenan.fyi
- Git's database internals I: packed object store - The GitHub Bloggithub.blog
- How Does a Database Work?cstack.github.io
- Code Storage by the Pierre Computer Companycode.storage
- The Hitchhiker's Guide to Ethereum - Delphi Digitalmembers.delphidigital.io
- Merkle Tree | Brilliant Math & Science Wikibrilliant.org
- From monolith to Lakebase to LTAP: rethinking the database from storage up | Databricks Blogdatabricks.com
- Log Structured Merge Trees | Ben Stopfordbenstopford.com