Edit

Share via

Persistent graphs overview (preview)

Applies to: ✅ Microsoft Fabric ✅ Azure Data Explorer

Note

This feature is currently in public preview. Functionality and syntax are subject to change before General Availability.

Persistent graphs enable you to store, manage, and query graph data structures at scale. Unlike transient graphs created with the make-graph operator, persistent graphs are durable database objects that persist beyond individual query executions, providing enterprise-grade graph analytics capabilities.

Overview

Persistent graphs consist of two primary components:

This architecture provides both flexibility in defining graph schemas and efficiency in querying graph data at scale.

Key benefits

Persistent graphs offer significant advantages for enterprise-scale graph analytics:

  • Durable storage: Graph models and snapshots persist in database metadata for long-term availability
  • Scalability: Handle large graphs that exceed single-node memory limitations
  • Reusability: Multiple users and applications can query the same graph structure without reconstruction
  • Performance optimization: Eliminate graph construction overhead for repeated queries
  • Schema support: Structured definitions for different node and edge types with their properties
  • Version control: Multiple snapshots enable representation of graphs at different points in time

Graph models

A graph model defines the specifications of a graph stored in your database metadata. It includes:

  • Schema definition: Node and edge types with their properties
  • Data source mappings: Instructions for building the graph from tabular data
  • Labels: Both static (predefined) and dynamic (generated at runtime) labels for nodes and edges

Graph models contain the blueprint for creating graph snapshots, not the actual graph data.

Managing graph models

The following commands are available for managing graph models:

Command Description
.create-or-alter graph_model Creates a new graph model or alters an existing one
.show graph_model Displays details of a specific graph model
.show graph_models Lists all graph models in the database
.drop graph_model Removes a graph model

Graph snapshots

A graph snapshot is the actual graph instance materialized from a graph model. It represents:

  • A specific point-in-time view of the data as defined by the model
  • The nodes, edges, and their properties in a queryable format
  • A self-contained entity that persists until explicitly removed

Snapshots are the entities you query when working with persistent graphs.

Managing graph snapshots

The following commands are available for managing graph snapshots:

Command Description
.make graph_snapshot Creates a new graph snapshot from a graph model
.show graph_snapshot Displays details of a specific graph snapshot
.show graph_snapshots Lists all graph snapshots in the database
.drop graph_snapshot Removes a single graph snapshot
.drop graph_snapshots Removes multiple graph snapshots based on criteria

Workflow

The typical workflow for creating and using persistent graphs follows these steps:

  1. Create a graph model - Define the structure and data sources for your graph
  2. Create a graph snapshot - Materialize the graph model into a queryable snapshot
  3. Query the graph snapshot - Use KQL graph operators to analyze the graph data
  4. Manage lifecycle - Create new snapshots as needed and drop old ones

Querying persistent graphs

Once a graph snapshot is created, it can be queried using the graph function followed by other KQL graph operators:

graph("MyGraphModel")
| graph-match (n)-[e]->(m)
    project n, e, m

To query a specific snapshot, provide the snapshot name:

graph("MyGraphModel", "MyGraphSnapshot")
| graph-match (n)-[e]->(m)
    project n, e, m

The graph-match operator enables pattern matching and traversal operations, while graph-shortest-paths helps find optimal connections between entities. The graph-to-table operator converts graph results back to tabular format.

Key considerations

This section describes key considerations and current limitations of graph models and snapshots.

Snapshot limitations

Persistent graphs have the following limitations:

  • Regular database limit: Maximum of 5,000 graph snapshots per database
  • Free virtual cluster limit: Maximum of 500 graph snapshots per database
  • Snapshot creation time: Limited to 1 hour

Next steps