Skip to main content

In-Memory Fluree

You can run Fluree in memory for testing purposes. To run Fluree in-memory, you must be running only one single server (not in a transactor group).

To do so, you will need to specify the following config options:

  1. fdb-consensus-type as in-memory
  2. fdb-storage-type as memory.

Full-text indexing and full-text search are not available when running Fluree in-memory.

To try this out, we can clone down the transactor-group-examples repo.

git clone https://github.com/fluree/transactor-group-examples.git

Then you can navigate to the in-memory/ directory. You'll need to download any version of Fluree (0.13 and higher) and copy fluree_server.jar into in-memory/ folder: in-memory/fluree_server.jar.

Then, you can open a terminal window and navigate to in-memory/. Issue ./fluree_start.sh to start running Fluree in-memory.

When running Fluree in-memory, nothing is written to disk, and all data is lost when the instance is shut down.