Digital twins, on one substrate

Q: Can SurrealDB handle high-frequency telemetry alongside graph queries?

Yes. Time-series ingests at high throughput and is queried in the same language as the graph, so one query joins telemetry to the entities and relationships it describes.

Q: How do simulation and scenario branching work?

Twins fork state, test alternatives, and merge or discard the result. Flexible schema and point-in-time queries with VERSION on a versioned storage engine (experimental) support this without touching live state, and Agent Memory extends it to the facts an agent would learn during the simulation.

Q: What about edge and air-gapped deployments?

SurrealDB is a single Rust binary that runs embedded, so a twin on a remote site, an aircraft or a disconnected facility runs offline with full ACID guarantees; replication to a central instance runs through your application. See /solutions/defence-and-aerospace and /solutions/energy-and-manufacturing.

Q: Where can I read more?

A full white paper on digital twins and the data backbone is available.

Start free with SurrealDB Talk to us

Entities. Every asset and its relationships modelled natively as one graph, with configuration as documents.

Telemetry. Time-series beside the graph in one transactional engine, joined to its assets in one query.

Agents. What an agent learns about a machine is stored against the machine, in the twin's own graph.

01 |RELATIONSHIPS + TELEMETRY

One graph for assets and their world. Relationships, telemetry and documents in one query.

1-- Find turbines within 50km of a depot, with their upstream dependencies
2-- and last five temperature readings, in one statement.
3SELECT
4    id, name, location,
5    ->depends_on->asset.{ id, name, status } AS upstream,
6    (SELECT value, recorded_at FROM sensor_reading
7       WHERE asset_id = $parent.id AND metric = "temperature"
8       ORDER BY recorded_at DESC LIMIT 5) AS recent_temps
9FROM asset
10WHERE kind = "turbine"
11  AND geo::distance(location, (-1.6, 53.8)) < 50000;

02 |REWIND, REPLAY, FORK

Versioned state, non-destructive history. Any moment, read back as a query.

SELECT ... VERSION reads a record as it stood at any moment on a versioned store. Add valid_from and valid_until to track when a fact held in the world, and scenario branching becomes a query rather than orchestration.

03 |AGENTS INSIDE THE TWIN

Agent Memory on the same engine. The agent's view of the twin and the twin itself share a transaction.

Typed relationships

What an agent learns about a machine is stored against the machine, in the twin's own graph.

Temporal state and provenance

Every remembered fact carries when it held and where it came from, beside the twin's versioned records.

Calibrated trust

Memory and twin share one transaction boundary, so the agent never acts on a state the twin has moved past.

Explore Agent Memory

Your questions,answered

How does SurrealDB compare to Neo4j for digital twin workloads?

A digital twin is rarely only a graph: telemetry is time-series, configuration is document-shaped, semantic search needs vectors, locations are geospatial, scenarios need versioned state. Neo4j is a dedicated graph database, so those become further systems joined through middleware. SurrealDB holds all of them in one engine and one query language.

Can SurrealDB handle high-frequency telemetry alongside graph queries?

How do simulation and scenario branching work?

What about edge and air-gapped deployments?

Where can I read more?

THE PLATFORM

Everything an application and its agents know. Five surfaces, one engine.

Database

Document, graph, vector, time-series and relational in one engine.

Five data models as dotted tiles: documents, graph, vector, time-series and relational

Agent Memory

What an agent learns, with its source and its time, in the same engine.

A timeline of remembered facts, each with its source

Cloud

Dedicated clusters by region, or serverless on SurrealDB Earth.

Clusters in three regions on a world map, each running or scaling

Studio

Query, explore and design the schema from the browser.

MCP

Every model that speaks MCP reaches the database and the memory directly.

Three models connected through MCP to the database and Agent Memory

IN PRODUCTION

Trusted at scale. Samsung, Nvidia, Verizon, Tencent, and Walmart run on SurrealDB.

14,000+

Developers building on SurrealDB Cloud

4M+

Developers building on SurrealDB worldwide

FROM THE TEAMS

SurrealDB gives us a foundation where we can unify semantic search, knowledge graphs, and AI-driven decision making without stitching together multiple systems. Collapsing responsibility into SurrealDB has become our default engineering posture.

Justin Foley

VP of Engineering, Later

GET STARTED

Build your digital twin on one backbone. Entities, relationships, telemetry, vectors and the agents that reason over them, in one engine.

Start free with SurrealDB Read the white paper