Redis

Featureform supports Redis as an Inference Store and a Vector DB

​

Implementation

In the inference store configuration, one Redis hash is created per feature. It maps entities to their feature value. A metadata hash is also stored in Redis that allows Redis to maintain its own state. This is used in conjunction with Featureform’s Etcd service to achieve consistency between the two.

​

Configuration

First we have to add a declarative Redis configuration in Python. In the following example, only name is required, but the other parameters are available.

import featureform as ff

ff.register_redis(
    name = "redis",
    description = "Example inference store",
    team = "Featureform",
    host = "0.0.0.0",
    port = 6379,
    password = "",
    db = 0,
)

client.apply()

Once our config file is complete, we can apply it to our Featureform deployment. Afterwards we can set it as the Inference Store or Vector DB when defining a feature or embedding respectively.

We can re-verify that the provider is created by checking the Providers tab of the Feature Registry or via the CLI.