Skip to main content
NetApp Console local deployment

Provision storage with AI-assisted (agentic) workflows in NetApp Console local deployment

Contributors netapp-tonias netapp-ml94669
PDFs

In NetApp Console local deployment, use AI-assisted (agentic) provisioning when you want to request storage conversationally, review a proposed plan, and approve the change. This approach is useful when you want faster requests with less UI navigation while still keeping provisioning aligned with storage classes and RBAC.

Note This documentation regarding connecting an LLM to the Console assistant to enable AI features is provided as a technology preview. With this preview offering, NetApp reserves the right to modify offering details, contents, and timeline before General Availability.
Required access roles

Super admin, Organization admin, Folder or fleet admin, or Storage admin. Your AI-assisted actions are limited by the same role-based access as UI-based actions. Learn about access roles.

Agentic provisioning requires an LLM integration to be configured and available in your NetApp Console local deployment environment.

Before you begin

  • Ensure LLM integration is configured and enabled for your NetApp Console local deployment.

  • Ensure the target fleet contains discovered storage systems.

  • Ensure storage classes are configured and associated with the fleet so the assistant can propose class-aligned plans.

Steps (to be revisited)

  1. Open the NetApp Console assistant.

  2. Describe what you need in plain language (for example, protocol, size, environment, performance intent, and target fleet if applicable).

  3. Review the proposed plan, including:
    The selected fleet and target system placement (if applicable)
    The storage class selected or recommended
    Key settings that will be applied (capacity behavior, protection schedule, security requirements, and any QoS-related intent if shown)

  4. Confirm or adjust the plan.
    If the plan selects an unexpected class or placement, specify the desired class, fleet, or constraints and regenerate the plan.

  5. Approve execution to provision the workload.

  6. Verify the created resource in the storage inventory and confirm it is associated with the intended storage class for ongoing drift monitoring.