Cloudera Docs

Creating a Cloudera AI Registry with the UI

Before you can start using Cloudera AI Registry you must create a Cloudera AI Registry for your environment.

  1. In the Cloudera console, click the Cloudera AI tile. The Cloudera AI Workbenches page displays.
  2. Click AI Registries in the left navigation pane.
  3. Click the Create AI Registry button. Cloudera AI displays the Create AI Registry dialog box.
  4. Select your environment from the Environment Name field.
  5. Enter the following Ozone S3 configuration information.

    • Ozone Access Key

    • Ozone Secret Key

    • Ozone Bucket

    • Ozone Endpoint

  6. Click the Create button. You can now use Cloudera AI Registry to register, track, and deploy your models.
  1. In the Cloudera console, click the Cloudera AI tile. The Cloudera AI Workbenches page displays.
  2. Click AI Registries in the left navigation pane.
  3. Click the Create AI Registry button. Cloudera AI displays the Create AI Registry dialog box.
  4. Select your environment from the Environment Name field. This is a mandatory field.
  5. Enter the following S3 compatible object storage configuration information for artifact storage and proper authentication.

    • The mandatory S3 Access Key

      The S3 compatible access key used for authentication grants read or write access to the specified S3 bucket. The credentials must have the minimum required permissions, including GetObject, PutObject, DeleteObject, and ListBucket, and must be an alphanumeric string.

    • The mandatory S3 Secret Key

      The S3 compatible secret key for authentication is used in conjunction with the S3 Access Key to enable secure authentication to S3 storage.

    • The mandatory S3 Bucket

      The S3 Bucket field specifies the name of the S3 bucket where AI artifacts, such as models, datasets, and metadata, are stored. The bucket name must adhere to S3 naming conventions, so it must be in lowercase, contain no spaces, and be between 3 and 63 characters in length.

      Cloudera AI Registry cannot be created in case the bucket name contains a back slash, that is '/'. When creating a Cloudera AI Registry instance, the system now performs a validation to ensure that.

    • The optional S3 Endpoint

      This field specifies the custom S3-compatible endpoint URL, allowing integration with S3-compatible storage services. The URL must be complete and include the protocol information.

    • The optional S3 Region

      The region must match the region where the bucket was created.

  6. Click the Create button. You can now use Cloudera AI Registry to register, track, and deploy your models.