Red Hat OpenShift AI Feature Store

What Red Hat OpenShift AI Feature Store is

Red Hat OpenShift AI Feature Store helps teams reuse machine-learning features for model training and inference. Teams can push features to the store and retrieve them as needed. It is based on the open-source Feast project and runs within self-managed OpenShift AI, which requires Red Hat OpenShift to be installed first. A cluster administrator enables the feature through the OpenShift AI Operator and creates an instance in a data science project.

The default local configuration uses a SQL registry or local file, a Parquet offline store, and a SQLite online store. The online container exposes a REST API inside the OpenShift cluster. The configuration documentation also covers a PostgreSQL online store and an S3 registry, and points to Feast for other storage options. The Feature Store supports online and offline stores, point-in-time joins, feature monitoring, and both serving modes.

Red Hat describes OpenShift AI as a platform for data scientists, AI engineers, developers, and IT operations teams. Its security positioning includes deploying models across on-premise, edge, and disconnected environments. OpenShift AI subscriptions offer Standard (8x5) or Premium (24x7) support SLAs. Pricing is on request, and Red Hat offers a 60-day self-supported OpenShift AI Self-Managed trial.

Who it suits

It suits data scientists, AI engineers, developers, and IT operations teams who need reusable features for training and inference in a self-managed OpenShift AI environment. It is also relevant to teams working across on-premise, edge, or disconnected deployments.

Limits

The OpenShift AI Self-Managed trial lasts 60 days and is self-supported. The listed subscriptions require Red Hat OpenShift; pricing is on request.

Getting started

1. Install Red Hat OpenShift.
2. Start the 60-day self-supported OpenShift AI Self-Managed trial, if appropriate.
3. Have a cluster administrator enable Feature Store through the OpenShift AI Operator.
4. Create a Feature Store instance in a data science project.
5. Configure the local stores or use the documented PostgreSQL online store and S3 registry options.

Verdict

Choose Red Hat OpenShift AI Feature Store if your team needs reusable machine-learning features within a self-managed OpenShift AI deployment. Teams seeking a priced-up-front option or an environment without Red Hat OpenShift should look elsewhere.