Chapter 1. Managing cluster resources
1.1. Configuring the default PVC size for your cluster
To configure how resources are claimed within your OpenShift Data Science cluster, you can change the default size of the cluster’s persistent volume claim (PVC) ensuring that the storage requested matches your common storage workflow. PVCs are requests for resources in your cluster and also act as claim checks to the resource.
Prerequisites
- You have logged in to Red Hat OpenShift Data Science.
Changing the PVC setting restarts the Jupyter pod and makes Jupyter unavailable for up to 30 seconds. As a workaround, it is recommended that you perform this action outside of your organization’s typical working day.
Procedure
-
From the OpenShift Data Science dashboard, click Settings
Cluster settings. - Under PVC size, enter a new size in gibibytes. The minimum size is 1 GiB, and the maximum size is 16384 GiB.
- Click Save changes.
Verification
- New PVCs are created with the default storage size that you configured.
Additional resources
1.2. Restoring the default PVC size for your cluster
To change the size of resources utilized within your OpenShift Data Science cluster, you can restore the default size of your cluster’s persistent volume claim (PVC).
Prerequisites
- You have logged in to Red Hat OpenShift Data Science.
- You are part of the OpenShift Data Science administrator group in OpenShift Container Platform.
Procedure
-
From the OpenShift Data Science dashboard, click Settings
Cluster settings. - Click Restore Default to restore the default PVC size of 20GiB.
- Click Save changes.
Verification
- New PVCs are created with the default storage size of 20 GiB.
Additional resources
1.3. Enabling GPU support in OpenShift Data Science
Optionally, to ensure that your data scientists can use compute-heavy workloads in their models, you can enable graphics processing units (GPUs) in OpenShift Data Science. To enable GPUs on OpenShift, you must install the NVIDIA GPU Operator. As a prerequisite to installing the NVIDIA GPU Operator, you must install the Node Feature Discovery Operator. For information about how to install these operators, see NVIDIA GPU Operator on Red Hat OpenShift Container Platform in the NVIDIA documentation.
Follow the instructions in this chapter only if you want to enable GPU support in an unrestricted self-managed environment. To enable GPU support in a disconnected self-managed environment, see Enabling GPU support in OpenShift Data Science instead.
1.4. Allocating additional resources to OpenShift Data Science users
As a cluster administrator, you can allocate additional resources to a cluster to support compute-intensive data science work. This support includes increasing the number of nodes in the cluster and changing the cluster’s allocated machine pool.
For more information about allocating additional resources to an OpenShift Container Platform cluster, see Manually scaling a compute machine set.
