terminusdb/terminusdb

TerminusDB is a distributed, collaborative database designed for building, sharing, versioning, and reasoning on structured data.

What it solves

TerminusDB is a distributed database designed to act as a "git for data," solving the problem of managing complex, linked data in a way that is collaborative, versioned, and queryable. It allows users to maintain a system of record for semantic knowledge graphs where changes can be tracked, branched, and merged like source code.

How it works

It links JSON and JSON-LD documents into a semantic knowledge graph. The system implements a revision control model using commits, diffs, and push/pull/clone operations to communicate changes between nodes. It provides multiple query interfaces, including a REST API, GraphQL, and WOQL (a Datalog logic engine) for goal-seeking and path queries. Version 12 introduces high-precision mathematics (xsd:decimal) and Allen Integer Algebra for temporal reasoning over ISO8601 date and time data.

Who it’s for

It is built for developers and organizations requiring a system of record for linked data, specifically those in industrial and financial use cases needing high precision math and temporal reasoning. It is also suited for ML/Ops and AI use cases that require deterministic symbolic AI processing.

Highlights

  • Git-like Versioning: Full revision control with commits, diffs, and time-travel queries to access any previous state.
  • Knowledge Graph Integration: Combines document-based storage (JSON) with the power of a semantic knowledge graph.
  • Advanced Reasoning: Built-in Datalog logic engine for unification and path queries, along with temporal interval classification.
  • Multi-interface Access: Supports GraphQL, REST, and WOQL for flexible data retrieval and deep link discovery.
  • High Precision: Supports arbitrary rational precision for decimals and temporal reasoning.

Related

  • Project
  • Project
  • Project
  • Project
  • Project