Digital twins, on one substrate

Q: How does SurrealDB compare to Neo4j for digital twin workloads?

Neo4j is a dedicated graph database. A digital twin is rarely only a graph: telemetry is time-series, configuration is document-shaped, semantic search needs vectors, asset locations are geospatial, scenario branching needs versioned state. With Neo4j, those become separate systems, joined through middleware. SurrealDB handles all of them natively, in one transactional engine and one query language. For digital twin workloads, this collapses several data layers into one.

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

Yes. SurrealDB ingests time-series data at high throughput and exposes it in the same query language as the graph, so a single query can join telemetry to the entities it describes and the relationships those entities sit in. There is no separate time-series store to keep in sync.

Q: How do simulation and scenario branching work?

Twins need to fork state, test alternatives, and either merge or discard the result. SurrealDB's flexible schema and versioned-record queries support this without corrupting the live state. Spectron extends the pattern to agent memory: a scenario can include the facts an AI agent would learn during the simulation, with the same provenance and supersession behaviour as the production substrate.

Q: What about edge and air-gapped deployments?

SurrealDB compiles to a single Rust binary with no external dependencies and runs embedded. Twins on remote sites, on aircraft, or in disconnected facilities can run a local instance and synchronise with a central one when connectivity returns. 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 being prepared for publication.

Entities, relationships, telemetry, and the agents that reason over them, all in one transactional engine.

RELATIONSHIPS + TELEMETRY

One graph for assets, dependencies, and the world they live in

A digital twin is a graph of entities - machines, suppliers, sites, people, vehicles - connected by typed relationships such as depends_on, supplies, operated_by, located_at. Around those entities sits telemetry, configuration, documents, and embeddings. In SurrealDB, all of these are first-class. A single SurrealQL query can walk the graph, filter by attribute, fetch the most recent sensor readings, and apply a geospatial predicate.

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;

REWIND, REPLAY, FORK

Versioned state and non-destructive history

Twins evolve. Equipment is replaced, configurations change, beliefs about a fault mode are revised when better evidence arrives. SurrealDB's VERSION clause lets you query a record as it existed at a given moment, so past state is reconstructed from the substrate itself rather than from bolt-on audit tables. Combine that with application-defined valid_from and valid_until fields to track when a fact held in the world - independent of when the database learnt about it - and scenario branching, what-if simulation, and forensic review of past decisions become queries rather than orchestration.

AGENTS INSIDE THE TWIN

Agent memory on the same engine

When AI agents operate inside a digital twin, they accumulate their own memory: what they have been told, what they have decided, what they now believe about the assets they reason over. That memory needs the same properties as the twin itself: typed relationships, temporal state, provenance, calibrated trust. Spectron provides this as a stateless application tier on the same SurrealDB substrate, with every fact carrying its source, every retrieval and decision recorded as a graph-resident trace, and contradictions surfaced as explicit uncertainty rather than silently overwriting. The agent's view of the twin, and the twin itself, share a transaction boundary. Learn more about Spectron at /platform/spectron.

FREQUENTLY ASKED QUESTIONS