Featureform supports Cassandra as an Inference Store.


A Cassandra table is created for every feature. All these tables are part of a keyspace with replication set to 3 by default. A metadata table exists within the keyspace as well to allow the provider to keep track of its own state. Featureform's scheduler aims to achieve consistency between Cassandra's internal state with the user's desired state as specified in the metadata service.


First we have to add a declarative Cassandra configuration in Python. In the following example, only name is required, but the other parameters are available.
import featureform as ff
name = "cassandra",
description = "Example inference store",
team = "Featureform",
host = "",
port = 9042,
username = "cassandra",
password = "cassandra",
consistency = "THREE",
replication = 3
Once our config file is complete, we can apply it to our Featureform deployment
featureform apply --host $FEATUREFORM_HOST
We can re-verify that the provider is created by checking the Providers tab of the Feature Registry.