Skip to main content

Create a Workbench compute pool

Add a compute pool to a Kubernetes cluster for Workbench sessions, services, training jobs, and endpoints.

Create a Workbench compute pool

Compute pools provide the Kubernetes node capacity that Workbench sessions, services, training jobs, and endpoints run on. This guide walks through creating a new pool from the Workbench interface.

Add a compute pool to a Kubernetes cluster for Workbench workloads.

TimeAbout 10 minutes
You will needA Kubernetes cluster in ready status, workspace admin permissions
OutcomeA compute pool ready for Workbench workloads

Open Pools

Click Workbench in the left sidebar, then click Compute > Pools.

Start pool creation

Click New Compute Pool.

Configure the pool

Set the following fields:

FieldRequiredWhat it does
NamerequiredUnique identifier for the pool, for example workers-cpu-small.
Min SizerequiredMinimum number of nodes. Set to 0 to allow scale-to-zero.
Max SizerequiredMaximum number of nodes for autoscaling.
Instance TyperequiredCloud machine type selected from the instance browser.
WorkloadrequiredWorkload tag such as workspace, training, inference, cloud-shell, pvc-viewer, or remote-desktop.
Root disk size (GB)requiredBoot disk size in GiB. Default is 30.
Secondary NetworksoptionalAdditional networks available to the pool.
GPU compute pooloptionalEnables GPU fields for GPU type, count, and vendor.

Create

Click Create.

Success looks like this: the compute pool is created and enters a ready state, available for session or job scheduling.

Set Min Size to 0 for scale-to-zero behavior. Nodes are provisioned on demand and terminated when idle. This reduces costs for bursty workloads but adds startup latency.

Verify the pool

The new pool appears in the Pools list with a node count and status. Sessions and training jobs can now target this pool when selecting compute resources.

Ask AI
Ask a question about Vantage Compute...