A relational database with SQL support built on FoundationDB
Java
677
5,188 commits
updated Oct 19, 2026
The FoundationDB Record Layer is a Java library that provides a record-oriented data store and a relational database interface built on top of the FoundationDB key/value store. Applications access it through a SQL interface over JDBC or a lower-level, record-oriented Java API. Its type system extends the standard relational types with user-defined nested structs, arrays, and fixed-dimension vectors for ML embeddings and similarity search. Reusable schema templates enable massively multi-tenant architectures whose shared schemas can safely evolve over time. An intelligent query planner handles joins, aggregation, grouping, sorting, and correlated subqueries, backing them with rich, incrementally-maintained indexes and favoring stream-based processing over in-memory buffers. Designed for distributed, stateless deployments with millisecond-level store initialization and query execution, it scales to thousands of independent database instances while inheriting the full ACID transactional semantics of FoundationDB.
1,674 followers · starred Jan 2019
190 followers · starred Jan 2019
657 followers · starred Sep 2023
354 followers · starred Jun 2021
A relational database with SQL support built on FoundationDB
Java
677
5,188 commits
updated Oct 19, 2026
The FoundationDB Record Layer is a Java library that provides a record-oriented data store and a relational database interface built on top of the FoundationDB key/value store. Applications access it through a SQL interface over JDBC or a lower-level, record-oriented Java API. Its type system extends the standard relational types with user-defined nested structs, arrays, and fixed-dimension vectors for ML embeddings and similarity search. Reusable schema templates enable massively multi-tenant architectures whose shared schemas can safely evolve over time. An intelligent query planner handles joins, aggregation, grouping, sorting, and correlated subqueries, backing them with rich, incrementally-maintained indexes and favoring stream-based processing over in-memory buffers. Designed for distributed, stateless deployments with millisecond-level store initialization and query execution, it scales to thousands of independent database instances while inheriting the full ACID transactional semantics of FoundationDB.
1,674 followers · starred Jan 2019
190 followers · starred Jan 2019
657 followers · starred Sep 2023
354 followers · starred Jun 2021