GitHub - JetBrains/youtrackdb: YouTrackDB is a general-use object-oriented graph database with storage format native to handle graph relations. YouTrackDB supports Gremlin queries and ACID transactions. YTDB supports both embedded and client-server deployment models.

4 min read Original article ↗

License official JetBrains project
Bluesky Zulip Medium Reddit

Issue tracker | Getting started | Documentation | Daily LDBC benchmarks

Join our Zulip community!

If you are interested in YouTrackDB, consider joining our Zulip community. Tell us about exciting applications you are building, ask for help, or just chat with friends 😃

What is YouTrackDB?

YouTrackDB is a general-purpose object-oriented graph database that supports both embedded and client-server deployment models. YouTrackDB is being supported and developed by JetBrains and is used internally in production.

YouTrackDB's key features are:

  1. Fast data processing: Links traversal is processed with O(1) complexity. There are no expensive run-time JOINs.
  2. Object-oriented API: This API implements rich graph and object-oriented data models. Fundamental concepts of inheritance and polymorphism are implemented on the database level.
  3. Snapshot isolation by default: All transactions run under snapshot isolation. Each transaction sees a stable snapshot of the database as of its start time, eliminating dirty reads, non-repeatable reads, and phantom reads.
  4. Implementation of TinkerPop API and Gremlin query language: You can use both Gremlin query language for your queries and TinkerPop API out of the box. Support of GQL with seamless integration with Gremlin is in progress. For maximum query performance, we suggest using YQL for initial data prefetching.
  5. YQL (YouTrackDB Query Language): A SQL-based query language with extensions for graph functionality. YQL uses intuitive dot notation for link traversal instead of JOINs, supports the powerful MATCH statement for graph pattern matching, and includes automatic index usage for query optimization.
  6. Scalable development workflow: YouTrackDB works in schema-less, schema-mixed, and schema-full modes.
  7. Strong security: A strong security profiling system based on user, role, and predicate security policies.
  8. Encryption of data at rest: Optionally encrypts all data stored on disk.
  9. Online incremental and full backups: Incremental and full backups run while the database keeps serving reads and writes, requiring only brief internal pauses to establish a consistent recovery point.
  10. Multiple databases per instance: A single YouTrackDB instance can host and operate several independent databases simultaneously — each with its own storage, schema, and security — giving you clean data separation through one unified API.

Easy to install and use

YouTrackDB can run on any platform without configuration and installation.

If you want to experiment with YouTrackDB, please check out our REPL console.

docker run -it youtrackdb/youtrackdb-console

To install YouTrackDB as an embedded database, add the following dependency to your Maven project:

<dependency>
  <groupId>io.youtrackdb</groupId>
  <artifactId>youtrackdb-embedded</artifactId>
  <version>0.5.0-SNAPSHOT</version>
</dependency>

The youtrackdb-embedded artifact is a shaded uber-jar that relocates all third-party dependencies (Guava, Jackson, Groovy, etc.) under com.jetbrains.youtrackdb.shade, so they won't conflict with versions used by your application.

You also need to add a YTDB snapshot repository to your Maven pom.xml file:

<repositories>
  <repository>
    <name>Central Portal Snapshots</name>
    <id>central-portal-snapshots</id>
    <url>https://central.sonatype.com/repository/maven-snapshots/</url>
    <releases>
      <enabled>false</enabled>
    </releases>
    <snapshots>
      <enabled>true</enabled>
    </snapshots>
  </repository>
</repositories>

or in case of Gradle:

dependencies {
    implementation 'io.youtrackdb:youtrackdb-embedded:0.5.0-SNAPSHOT'
}

and

repositories {
    maven {
        url = uri("https://central.sonatype.com/repository/maven-snapshots/")
        mavenContent {
            snapshotsOnly()
        }
    }
}

If you want to work with YouTrackDB server, you can start it using the Docker image:

docker run -p 8182:8182 \
 -v $(pwd)/secrets:/opt/ytdb-server/secrets \
 -v $(pwd)/databases:/opt/ytdb-server/databases \
 -v $(pwd)/conf:/opt/ytdb-server/conf \
 -v $(pwd)/log:/opt/ytdb-server/log \
 youtrackdb/youtrackdb-server

and provide root password for the database in the secrets/root_password file.

YouTrackDB requires at least JDK 21.

To learn how to use YouTrackDB, see the Getting Started guide.

For more examples covering both server and embedded deployments, check out the examples project.

Contributing

To understand how the query engine itself works, read Inside the YouTrackDB Query Engine, a deep-dive book explaining how YouTrackDB compiles and runs MATCH/YQL queries — from SQL text through the parser, pattern graph, cost-based planner, execution steps, and traversers — starting at Chapter 1 — Why a Graph Database Has Its Own Query Engine.

We develop YouTrackDB with the pi coding agent and strongly recommend you use it too. Our ytdb-slate extension ships Slate, a new agentic orchestration architecture from a Random Labs technical report. An orchestrator plans while worker threads execute, under mandatory design and adversarial reviews before implementation. It is pinned in our pi settings, so a clone uses the version we do. Our own coding, test, and verification policy layers on top — see docs-internal.