Home > Workload Solutions > Container Platforms > Red Hat OpenShift Container Platform > Guides > Implementation Guide—Red Hat OpenShift Container Platform 4.14 on AMD-powered Dell Infrastructure > OpenShift AI
Red Hat OpenShift AI is a platform for data scientists and developers of artificial intelligence (AI) applications. The platform provides a fully supported environment that lets you rapidly develop, train, test, and deploy machine learning models on-premises.
Ensure that you have:
To enable GPU support on an OpenShift cluster, follow the instructions in NVIDIA GPU Operator on Red Hat OpenShift Container Platform.
To deploy OpenShift AI:
When the installation is complete, Ready for Use is displayed.
The Status changes to Ready.
From the OpenShift Container Platform web console:
You are redirected to the rhods-dashboard login page.
3. Log in with your OpenShift credentials.
The RHODS dashboard opens, as shown in the following figure:
Figure 39. RHODS dashboard
4. To ensure that your data scientists can use compute-heavy workloads in their models, enable graphics processing units (GPUs) in OpenShift AI. See Enabling GPU support in OpenShift AI.
To view the reset migration-gpu-status instance, go to the the Instances tab on the Accelerator Profile CRD details page.
Figure 40. Accelerator Profiles
Creating a project lets you organize your data science work in one place. You can enhance the capabilities of your project by creating workbenches and adding storage, data connections, and model servers to your cluster.
To create a data science project:
Figure 41. Creating a Data Science project
A page opens where you can create workbenches, add cluster storage and data connections, import pipelines, and deploy models.
To create a project workbench:
The Create workbench page opens, as shown in the following figure:
Figure 42. Creating a data science project workbench
3. Provide the following values:
4. Configure the storage for your OpenShift AI cluster.
5. Click Create Workbench.
The status changes from “starting” to “running.”
Figure 43. Data Science project status
6. Click Open to get started on the project.
Figure 43. Sample project in the Jupyter Notebook console
Membership of the rhods-admins
admin group is required for access to the RHODS Settings panel.
To add a user to the rhods-admins
group:
kubeadmin
to the rhods-admins
group by running the following command:users:
- b64:kube:admin
Note: The OpenShift API specification mandates that usernames containing the colon punctuation mark are prefixed with b64: to be valid.
The Cluster Settings page is displayed on the RHODS dashboard, as shown in the following figure:
Figure 44. Cluster settings in the RHODS dashboard