Create a node

A privacy-preserving computation node stores raw data for a project. It connects to an organization's raw data to perform local computations. This topic describes how to create a privacy-preserving computation node on the Ant Privacy-Preserving Computation Platform.

Prerequisites

You have activated the Ant Privacy-Preserving Computation Platform. For more information about the activation process, see Activation guide.

Procedure

Follow these steps to create a node in the Management Console:

1. Log on to the Ant Privacy-Preserving Computation Platform.

2. In the navigation pane on the left, choose My Subscriptions > Privacy-Preserving Computation Nodes.

3. On the Node Management tab, click Create Node.

4. On the Create Node page, configure the parameters for the node.

   The following parameters are required to create a node:

   • Platform Configuration

     • Organization: The organization that the node belongs to.

     • Central Node: A central node acts as a central gateway for a project. When you activate the Ant Privacy-Preserving Computation Platform, a central node is automatically configured for you. Select the default value.

   • Organization Node Configuration

     • Organization Node Name: The name can contain Chinese characters, letters, digits, underscores (_), and hyphens (-). The name can be up to 64 characters long.

     • Organization Node Environment: Select the environment for the project in which the node will participate. You can select a production environment or a staging environment.

     • Organization Node Engine: Select the engine type based on the project type that the node will participate in. For more information about the relationship between project types and engine types, see the following table.

       Project Type	Engine Type
       Federated Modeling	Federated Learning
       Multi-party Secure Modeling	Training and Prediction
       Multi-party Security Analytics	Analytics
       Private Set Intersection	Analytics

       Note

       When you select both prediction and training engines, or both prediction and analytics engines, analytics or training tasks might compete for prediction service resources. This competition can cause the prediction service to become unstable. Select these combinations carefully. For more information about the resource configuration requirements for a single privacy-preserving computation node based on the organization node environment and engine, see Single-node resource configuration.

   • Network Configuration

     Network configuration is optional. You can configure the network information for the node after you deploy the client.

     • Communication Line: Select whether the node client will access the platform over the Internet or a leased line.

       • Internet: The client accesses the public IP address of the platform. This IP address is assigned by the central node.

       • Leased Line: The client accesses the private IP address of the platform. Contact the helpdesk to obtain the leased line address for the platform.

     • Organization Node Public Key: Leave this empty. After the node is created, you can upload the node's public key. For more information, see Install and deploy a node.

       Note

       The organization node public key is available only after the client is deployed. You do not need to configure this parameter now.

5. After configuring the node information, click Create in the lower-right corner.

6. On the Privacy-Preserving Computation Nodes page, confirm that the Node Status is Pending Deployment. This status indicates that the node was created successfully.

What to do next

Install and deploy a node