A new version, v2.6, of the NetApp DataOps Toolkit is now available. In addition to some minor enhancements, this new version adds two new MCP servers - one for orchestrating ONTAP volumes, and one for orchestrating persistent volumes and Jupyter workspaces in a Kubernetes environment.
What is an MCP Server?
MCP (Model Context Protocol) is a quickly emerging as a key building block in the Agentic AI space. So what exactly is it? In short, it is an open-source standard for connecting AI applications to external systems. AI applications, including standalone applications such as ChatGPT and custom applications that are developed using frameworks like LangGraph, can utilize MCP to access information and perform tasks using external systems and tools. MCP uses a client-server architecture, where the clients are AI applications and the servers are programs that connect the client applications with external systems and tools. MCP servers can either be local (on the same host as the client) or remote.
NetApp DataOps Toolkit MCP Server for ONTAP
The new version of the NetApp DataOps Toolkit introduces two MCP servers. The first of these is an MCP server for orchestrating volumes within ONTAP. This MCP Server enables AI applications to manage volumes and snapshots on an ONTAP system, including NetApp AFF and FAS appliances, Amazon FSx for NetApp ONTAP instances, and NetApp Cloud Volumes ONTAP instances. In this initial release, it only supports NFS volumes. This is a "local" MCP server, meaning that it needs to run on the same host as the client application. It communicates with remote ONTAP systems via REST API. The following tools are available as of the this initial release.
- Create Volume: Rapidly provision new data volumes with customizable configurations.
- Clone Volume: Create near-instantaneous, space-efficient clones of existing volumes using NetApp FlexClone technology.
- List Volumes: Retrieve a list of all existing data volumes, with optional space usage details.
- Mount Volume: Mount existing data volumes locally as read-only or read-write.
- Create Snapshot: Create space-efficient, read-only copies of data volumes for versioning and traceability.
- List Snapshots: Retrieve a list of all snapshots for a specific volume.
- Create SnapMirror Relationship: Set up SnapMirror relationships for efficient data replication.
- List SnapMirror Relationships: Retrieve a list of all SnapMirror relationships on the storage system.
For details on how to use this MCP server, refer to the documentation.
NetApp DataOps Toolkit for Kubernetes MCP Server
We have also released an MCP server for Kubernetes-based environments. This MCP Server enables AI applications to manage persistent volumes and Jupyter workspaces within a Kubernetes cluster, using the NetApp Trident CSI driver. This is also a "local" MCP server, meaning that it needs to run on the same host as the client application. It communicates with Kubernetes and Trident via the Kubernetes API. This MCP server is compatible with all NetApp storage platforms that are supported by Trident, including ONTAP, Amazon FSx for NetApp ONTAP, Azure NetApp Files, and Google Cloud NetApp Volumes. The following tools are available as of the this initial release.
Workspace Management Tools
Note: Any JupyterLab workspaces that is created by this MCP server will automatically be backed by NetApp storage.
- Create JupyterLab: Create a new JupyterLab workspace.
- Clone JupyterLab: Clone an existing JupyterLab workspace.
- List JupyterLabs: List all JupyterLab workspaces.
- Create JupyterLab Snapshot: Create a snapshot of a JupyterLab workspace.
- List JupyterLab Snapshots: List all snapshots of JupyterLab workspaces.
- Create Volume: Create a new volume.
- Clone Volume: Clone an existing volume.
- List Volumes: List all volumes.
- Create Volume Snapshot: Create a snapshot of a volume.
- List Volume Snapshots: List all snapshots.
For details on how to use this MCP server, refer to the documentation.
Demo with GitHub Copilot Chat in VS Code
GitHub Copilot Chat is a popular MCP client in the enterprise space. To give you an example of how our MCP servers work, we added the NetApp DataOps Toolkit MCP Server for ONTAP to a GitHub Copilot Chat instance running in VS Code and recorded a short demo. You can view the demo below to see our MCP server in action!
Further Reading
To learn more about the NetApp DataOps Toolkit, visit the project on GitHub. You can also find an overview of the MCP servers that are available with the NetApp DataOps toolkit here. If you would like to request a new tool or capability, we invite you to submit your request here.
