Module 2: Red Hat OpenShift AI configuration

Red Hat OpenShift AI running on Single Node OpenShift provides the complete AI/ML platform for model training, serving, pipelines, and monitoring. This module reviews the pre-configured Red Hat OpenShift AI environment optimized for edge deployments.

Learning objectives

By the end of this module, you will be able to:

  • Access the Red Hat OpenShift AI Dashboard

  • Navigate data science projects and understand how they organize AI/ML resources

  • Review DataConnections linking RHOAI to robot storage

Exercise 2.1: Access the Red Hat OpenShift AI web interface

The Red Hat OpenShift AI operator is installed and pre-configured in the Single Node OpenShift cluster.

Access the Red Hat OpenShift AI console

Access the Red Hat OpenShift AI dashboard:

  1. Open the RHOAI dashboard and login with credentials:

    https://rhods-dashboard-redhat-ods-applications.apps.cluster.example.com
    Credential Value

    Username

    admin

    Password

    password

  2. Click Allow selected permissions if prompted

  3. You’ll see the RHOAI landing page

RHOAI Dashboard landing page showing Data Science Projects and other sections

Verify

✓ Red Hat OpenShift AI Dashboard loads successfully

✓ The landing page shows the ai-edge-project pre-created

Exercise 2.2: Review data science project

Data Science Projects in Red Hat OpenShift AI organize users and AI/ML resources in a single namespace. For our battery monitoring solution, the ai-edge-project contains all necessary components: workbenches, data connections, model servers, and pipelines.

Access the ai-edge-project

  1. Ensure you’re in the Red Hat OpenShift AI Dashboard

  2. Click Projects in the left navigation

  3. Locate and click on ai-edge-project that has been pre-created

AI & Edge Data Science Project

The project dashboard shows multiple tabs for managing AI/ML resources.

Explore project components

Review the project dashboard tabs:

  • Overview - Project details and quick access to resources

  • Workbenches - JupyterLab development environments

  • Pipelines - Automated ML workflows

  • Deployments - Deployed model servers for inference

  • Connections - External storage and data sources

  • Permissions - User access control

These tabs organize all resources for the battery monitoring AI solution.

Verify

Confirm the project is properly configured:

✓ ai-edge-project appears in Data Science Projects list

✓ Project dashboard shows all tabs (Overview, Workbenches, Pipelines, Deployments, Connections)

Exercise 2.3: Review data connection

Data connections link Red Hat OpenShift AI to external object storage. The vehicle-models connection provides access to the robot’s MinIO storage where AI models are stored. The connection bridges the SNO training environment and robot’s edge storage.

Review the vehicle-models data connection

  1. Ensure you’re in the RHOAI Dashboard inside the ai-edge-project project

  2. Navigate to the Connections tab

  3. Locate the existing connection: vehicle-models

  4. Click on the three vertical dots (⋮) on the right and click Edit

    Connections tab showing vehicle-models DataConnection

Understand connection parameters

The vehicle-models connection is configured with:

Table 1. Data connection configuration
Parameter Value Purpose

Connection type

S3 compatible object storage - v1

Standard S3 protocol compatible with MinIO

Connection name

vehicle-models

Identifier for this data connection

Access key

minio

MinIO access credential

Secret key

minio123

MinIO access password

Endpoint

http://minio-microshift-vm.microshift-001.svc.cluster.local:30000

Robot’s MinIO storage service URL

Region

us-east-1

Default S3 region, required but unused by MinIO

Bucket

inference

Contains AI models in the robot

Close (x) the configuration view

Data connections are stored as Kubernetes Secrets. The secret contains base64-encoded S3 credentials:

  • AWS_ACCESS_KEY_ID - MinIO username

  • AWS_SECRET_ACCESS_KEY - MinIO password

  • AWS_S3_ENDPOINT - MinIO API endpoint

  • AWS_DEFAULT_REGION - S3 region

  • AWS_S3_BUCKET - Target bucket name

All values are base64-encoded for security. Red Hat OpenShift AI components automatically decode them when using the connection.

Verify

✓ vehicle-models data connection appears in Connections tab

✓ Data connection points to MicroShift MinIO service

✓ Data connection uses the inference bucket

Summary

You have reviewed the pre-configured Red Hat OpenShift AI environment on Single Node OpenShift:

✓ Red Hat OpenShift AI Dashboard - Web interface accessible and operational

✓ ai-edge-project - Data Science Project organizing all battery monitoring AI resources

✓ vehicle-models data connection - S3 connection linking SNO to robot’s MinIO storage

What You’ve Learned:

  • Red Hat OpenShift AI provides a complete AI/ML platform on SNO

  • Data Science Projects organize resources in namespaces

  • Data connections link Red Hat OpenShift AI to external storage (robot’s MinIO)

  • Secrets store S3 credentials securely

  • The SNO server can access robot storage over the factory network

The Red Hat OpenShift AI environment is ready for model training and automated retraining workflows.

Next, you’ll explore the pre-configured workbench where data scientists train battery health prediction models.

Navigate to Module 3: Model training.