Integrated Graph Database Features

High-performing, scalable, and secure Graph analysis

As a fully integrated feature of Oracle Database, Oracle Graph inherits security, scalability, and high-performing capabilities.

Oracle scored highest in the 2023 Gartner Critical Capabilities for Cloud Database Management Systems for Operational Use Cases.


Comprehensive Graph support

Oracle Graph supports both RDF (resource description framework) and property graphs on operational data. Developers can now easily build graph applications with SQL using existing SQL development tools and frameworks through SQL property graphs. The combination of AI Vector Search and RDF knowledge graph capabilities in Oracle Database makes it simple to extend retrieval-augmented generation (RAG) workflows with graph analysis, enabling improved search response accuracy—also known as GraphRAG.

RDF graphs

RDF graphs are designed to represent statements and are best for representing complex metadata and master data. Use Oracle Graph, which adheres to W3C standards, for linked data, data integration, and knowledge graphs.

Property graphs

Use property graphs to model data as vertex and edge relationships to query and analyze data based on those relationships.

SQL property graphs

Oracle Database 23ai is the first commercial database to implement the new SQL:2023 standard, making it simple for anyone with SQL knowledge to define and query graph models.

Read the SQL property graphs datasheet (PDF)


More than 80 out-of-box parallelized, in-memory algorithms

Oracle Graph includes more than 80 graph algorithms enabling you to perform ranking, community detection, pathfinding, link prediction, machine learning (ML), and more. You can also use the output as input to a machine learning process so your ML models can use the indirect relationships to improve the accuracy of the predictions. The algorithms are parallelized for scalability and performance against operational data.

Community detection algorithms

Includes: Strongly Connected Components, Weakly Connected Components, Label Propagation, Louvain, Conductance Minimization, Infomap, and Speaker-Listener Label Propagation.

Topology analysis algorithms

Includes: Conductance, Cycle Detection, Degree Distribution, Eccentricity, K-Core, LCC, Modularity, Reachability Topological Ordering, Triangle Counting, Bipartite Check, Partition Conductance, and Reachability

Ranking and walking algorithms

Includes: PageRank, Personalized PageRank, Degree Centrality, Closeness Centrality, Vertex Betweenness Centrality, Eigenvector Centrality, HITS, Minimum Spanning-Tree (Prim's), Breadth-First Search, Depth-First Search, Random Walk with Restart, Article Rank, and Harmonic Centrality.

Path-finding algorithms

Includes: Shortest Path (Bellman-Ford, Dijkstra, Bidirectional Dijkstra), Fattest Path, Compute Distance Index, Enumerate Simple Paths, Filtered and Unfiltered Fast Path Finding, Hop Distance, All Reachable Vertices and Edges, and Compute High-Degree Vertices.

Link prediction and other algorithms

Includes: Who-to-follow, SALSA, and Adamic-Adar Index.

Machine learning algorithms

Includes: DeepWalk, Supervised GraphWise, Unsupervised GraphWise, Pg2Vec, Matrix Factorization, and GNNExplainer.


Oracle Graph Studio in Autonomous Database

With Graph Studio, almost anyone can get started with graphs to explore relationships in data. Graph Studio removes barriers to entry by automating complicated setup and management, making data integration seamless, and by providing step-by-step examples for getting started, all while offering powerful algorithms, a speedy in-memory analytics server, and advanced visualization.

Read the ebook (PDF)

Graph Studio includes:

  • Support for SQL property graph, PGQL property graph, and RDF graph
  • Automated graph modeling
  • Extensive graph analytics and graph query support
  • Advanced notebooks and integrated visualization
  • Automated install, upgrade, and provisioning

Graph Studio is included at no additional cost in Autonomous Database Free Tier, Autonomous Data Warehouse Serverless, and Autonomous Transaction Processing Serverless.

For more information on Graph Studio, see the Graph Studio FAQ.

Graph modeling

Use an intuitive UI to create SQL property graph, PGQL property graph, and RDF graph. You can create a property graph from relational tables or from an RDF graph. Import RDF data and create an RDF graph or an RDF graph collection through a wizard.

Graph visualization

Visually explore and interact with a graph to discover patterns. You can annotate and save the graph to communicate your discoveries with others.

Notebooks

Improve productivity and team collaboration by developing, organizing, executing, and sharing code through an interactive, browser-based notebook with nine interpreters. You can also visualize results without using the command line or installing a separate tool.

Interpreters support

  • SQL to run SQL queries on graphs and tables as well as to create and insert data into graphs and tables.
  • pgql-rdbms to run PGQL queries in database.
  • pgql-pgx to run PGQL queries against the embedded Graph Server.
    • This requires the graph is loaded into memory through the UI or programmatically through the python-pgx or java-pgx paragraphs.
    • When algorithms are run from the python-pgx or java-pgx paragraphs, the results are reflected on the in-memory graph, so you would use this interpreter to query for those results.
  • python-pgx and java-pgx to write custom Python or Java code, including using the Python/Java APIs to load graphs to the embedded Graph Server and run graph algorithms.
  • custom-algorithms-pgx to write your own custom PGX graph algorithms.
  • sparql-rdf to write sparql queries on an RDF graph.
  • Markdown to add descriptions to your notebook.
  • Conda to install third-party libraries.

In-memory performance

Using Autonomous Database as its persistent data layer, Graph Studio moves graph data into an in-memory structure for fast and efficient analysis. Graph Studio automatically calculates required memory allocation, so you do not have to manually manage the required resources.


Graph Server and Client

Oracle Graph Server and Client enables developers, analysts, and data scientists to use graphs within Oracle Database. It may also be used as a user-managed graph environment with Oracle.

It includes a high-speed, in-memory, parallel server for property graph queries and analytics, and client components, such as command-line shells for working with the graph API, a plugin for SQLcl to run PGQL queries, and a graph visualization tool.

Download Oracle Graph Server and Client

Oracle scored highest in the 2023 Gartner Critical Capabilities for Cloud Database Management Systems for Operational Use Cases.

Oracle Graph Server and Client includes:

  • Graph Server
  • Graph Client
  • PGQL Plugin for SQLcl
  • Graph visualization application
  • Graph web apps

Oracle RDF Graph adapters and plugins enables using open source RDF clients and development frameworks with the RDF graph feature in Oracle Database. It includes support for RDF Graph adapter for Eclipse RDF4J and Apache Jena 3.12.0, Apache Jena Fuseki 3.12.0, and Protege Desktop.

Graph modeling

Use SQL or PGQL to write a Create Property Graph statement and create a property graph on your tables. This can be executed from any SQL- or PGQL-supported tool.

Graph visualization

Graph Server includes a Graph visualization application you can use to visually explore and interact with a graph to discover patterns. Graph Server also has REST API endpoints that you can use with the Oracle Graph Visualization Library to visualize graphs in your JavaScript application.