What it is

OrientDB is an open-source multi-model NoSQL DBMS written in Java under the Apache-2.0 license. It belongs to the database ecosystem as one product supporting graph, document, reactive, full-text, geospatial, and object-oriented concepts. It can run distributed in multi-master mode and offers SQL alongside ACID transactions.

The problem it addresses is storing and querying linked data without several specialized databases. A team can keep graph relationships, document records, full-text indexes, and geospatial information in one OrientDB deployment, then query through SQL or traverse through persistent pointers. The README says connections are managed as persistent pointers between records, so the system can avoid expensive runtime JOINs and traverse many records quickly. It supports schema-less, schema-full, and schema-mixed modes for unstructured, structured, or partially structured data.

Key capabilities

• OrientDB supports multiple data models in one product, including native graphs, documents, reactive queries, full-text search, and geospatial concepts.
• It uses persistent pointers between records instead of expensive runtime JOINs, enabling fast traversal of linked data.
• It provides a SQL query layer, making the database approachable for people already skilled in relational systems.
• It supports ACID transactions, schema-less, schema-full, and schema-mixed modes, plus security profiling based on users, roles, and predicate security.
• It can run distributed in multi-master mode and is written in Java for cross-platform deployment.

Who uses it and how

• Teams that need graph and document data together can store linked records and query them through SQL or graph traversal.
• Relational developers can use OrientDB as a NoSQL database without abandoning SQL as their primary query language.
• Java applications can run OrientDB on any platform that supports Java, using the database without configuration and installation according to the README.
• Distributed deployments can use OrientDB's multi-master capability when multiple nodes need to serve the same data.

Getting started

The README says OrientDB is written in Java and can run on any platform without configuration and installation, and it points users to the Getting Started documentation. No Docker image, package manager, or hosted service is named in the provided README excerpt.

When to use it — and when not to

OrientDB is useful when one database must handle graph, document, full-text, reactive, and geospatial models together, and when SQL familiarity matters. It is not a relational DBMS, despite SQL support, so teams requiring a traditional relational model should evaluate whether its NoSQL design fits their data. The provided facts list 360 open issues and do not mention storage or SMTP requirements, so self-hosters must operate the Java database, security, schemas, and distributed configuration themselves.

@inproceedings{DBLP:conf/bncod/0001DLT21,
  author    = {Daniel Ritter and
               Luigi Dell'Aquila and
               Andrii Lomakin and
               Emanuele Tagliaferri},
  title     = {OrientDB: {A} NoSQL, Open Source {MMDMS}},
  booktitle = {Proceedings of the The British International Conference on Databases
               2021, London, United Kingdom, March 28, 2022},
  series    = {{CEUR} Workshop Proceedings},
  volume    = {3163},
  pages     = {10--19},
  publisher = {CEUR-WS.org},
  year      = {2021}
}