Skip to main content

How Fluree Stores Data

Understanding how Fluree organizes and stores data will help you work with it effectively.

Ledgers​

A ledger is Fluree's container for data—similar to a database in relational systems. When you create a ledger, you're creating a place to store related data.


curl --location 'http://localhost:58090/fluree/create' \
--header 'Content-Type: application/json' \
--data '{
"ledger": "my-app/data"
}'

Entities​

Data in Fluree represents entities—things you want to record information about. An entity could be a person, a product, an event, or any concept relevant to your domain.

Entities are represented as JSON objects:


{
"@id": "ex:alice",
"name": "Alice",
"email": "alice@example.com"
}

The @id uniquely identifies the entity. Every entity should have one.

The Triple Model​

Under the hood, Fluree stores data as triples: subject-predicate-object statements.

The entity above becomes:

  • ex:alice → name → "Alice"
  • ex:alice → email → "alice@example.com"

This model enables flexible graph queries and relationships, but you interact with Fluree using familiar JSON.

Transactions and Commits​

When you insert or modify data, you create a transaction. Each transaction creates a new commit—an immutable snapshot of the change.


{
"ledger": "my-app/data",
"insert": {
"@id": "ex:alice",
"name": "Alice"
}
}

This append-only model means:

  • No data is ever overwritten
  • You can query any historical state
  • Changes are auditable

Next Steps​