Skip to main content

Private Cluster

Overview​

Private Cluster enables you to create a dedicated pool of GPU-backed compute resources that can be shared across multiple teams and projects. The cluster is billed at a fixed price, independent of actual GPU utilization, providing predictable costs and guaranteed capacity.

Once a Private Cluster is created, you can deploy Nodes, Inference Endpoints, Training Clusters, and Vector Databases within it without incurring additional charges.

Who Should Use a Private Cluster?​

Private Clusters are best suited for teams that:

  • Run long-running or predictable AI/ML workloads
  • Require guaranteed GPU availability
  • Want to share GPUs across multiple projects without repeated provisioning
  • Prefer fixed and predictable billing
  • Need controlled access through role-based permissions

When to Choose a Private Cluster?​

Choose a Private Cluster if you want to:

  • Avoid GPU shortages during peak demand
  • Eliminate per-resource billing for nodes, inference, and databases
  • Centrally manage GPU capacity across teams
  • Control GPU allocation without deleting workloads
  • Reduce operational overhead for large or growing AI teams

Key Benefits​

  • Fixed Pricing – Billing is independent of actual GPU usage
  • No Hidden Costs – No extra charges for deploying services inside the cluster
  • Guaranteed Capacity – GPUs are reserved exclusively for your organization
  • Flexible Allocation – Dynamically allocate and deallocate GPUs across projects
  • Secure Access Control – Governed by IAM roles and permissions

Creating a Private Cluster​

Step-by-Step Guide​

  1. Navigate to Private Cluster from the sidebar under Products.
  2. Click Create Private Cluster.
  3. Select the required compute configuration and choose a pricing option:
    • Hourly – Flexible, pay-as-you-go pricing
    • Committed – Lower cost for predictable workloads with a fixed commitment period
  4. Enter the Cluster Name and specify the Node Count.
  5. Click Create.

Commitment Options (Committed Plans Only)​

If you select a committed plan, choose what happens after the commitment period ends:

  • Auto-Renew – Continue the commitment automatically
  • Switch to Hourly Billing – Retain the cluster with hourly pricing
  • Auto-Terminate – Automatically delete the cluster
Note

Committed plans provide cost benefits for steady workloads. Any additional nodes added later are billed at hourly rates.

Understanding Roles and Permissions​

Private Cluster access is controlled using IAM roles, ensuring that only authorized users can manage cluster capacity.

How Access Works​

  • Admins and Owners manage cluster creation and capacity.
  • Team Leads and Project Leads manage GPU usage within their scope.
  • Team Members typically have read-only visibility unless explicitly granted permissions.

Role-Based Access Matrix​

RoleView ClusterCreate / Update ClusterAllocate NodesDeallocate NodesScope
Admin / OwnerYesYesYesYesCluster (CRN)
Team LeadYesNoNoYesAll projects in team
Project LeadYesNoNoYesAssigned project only
Team MemberYes (Read-only)NoNoYes (if permitted)As per IAM policy

Common Scenarios​

  • I am a Team Lead and want to free GPUs from a project → Allowed
  • I am a Project Lead and want to resize the cluster → Not allowed
  • I am a Member and want to view cluster usage → Allowed

How Node Allocation Works​

Node Lifecycle States​

Each node in a Private Cluster exists in one of the following states:

  • Free – Available for allocation
  • Allocated – Assigned to a project
  • Occupied – Actively running workloads (Nodes, Inference, Training, Databases)

Understanding these states helps you manage capacity without disrupting running services.