Cloudera Docs

Deploying a CDV application in CML

Learn how to deploy Cloudera Data Visualization (CDV) in Cloudera Machine Learning (CML).

Creating a CML project with CDV Runtime

Learn how to create a Cloudera Machine Learning (CML) project with Cloudera Data Visualization (CDV) Runtime as the default runtime.

If you know that your project is going to be running CDV, you can add the Cloudera Data Visualization Runtime when setting up the project.
  1. Click Projects on the left sidebar of Cloudera Machine Learning data service.
  2. Click New Project.
  3. Enter a project name.
  4. Optional: You can add a description for your project.
  5. Optional: You can also select the visibility of the project.
  6. Under Initial Setup, you can either create a blank project, or select a source for your project files.
  7. Set Cloudera Data Visualization Runtime for your project in the Runtime setup section.
    You can use the Basic view to select the Cloudera Data Visualization Runtime and add all of its Standard Edition ML Runtimes for that kernel. In the Advanced view, you can add ML Runtimes based on more detailed Editor, Kernel, Edition, and Version criteria.
  8. Click Create Project.

After the project is created, you can start creating your application. If you added Cloudera Data Visualization Runtime as the only runtime during project set-up, it will be the only available runtime when creating your application. You can select which edition you want to use for the application.

Adding CDV Runtime to a CML project

Learn how to add Cloudera Data Visualization (CDV) Runtime to an existing Cloudera Machine Learning (CML) project.

You need to add Cloudera Data Visualization Runtime to your project if the workspace of your project is not set to use this CDV Runtime by default.
  1. Click Projects on the left sidebar of Cloudera Machine Learning data service.
  2. Click the project where you want to add Cloudera Data Visualization Runtime.
  3. Open Project Settings from the left navigation bar and click the Runtime/Engine tab.
  4. Click Add Runtime.
    The Add new runtime to project modal window opens.
  5. Select Cloudera Data Visualization from the Kernel drop-down list.
  6. Select the edition of the CDV you want to add to your project.
    The selected edition automatically defines the version of the runtime variant.
  7. Click Submit.
You should now be able to select the Cloudera Data Visualization Runtime when creating a CDV application.
Create a CDV application.

Creating CDV application in CML

Learn how to create a Cloudera Data Visualization (CDV) application in Cloudera Machine Learning (CML).

Ensure that CDV Runtime is available the CML project where you want to create the CDV application.

For more information about ML Runtimes, see Managing ML Runtimes and Using Runtime Catalog.

  1. Go to your CML project's Overview page.
  2. On the left sidebar, click Applications.
  3. Click New Application.
  4. Provide the following details for your new application:
    • Name – Enter a name for the application.
    • Subdomain – Enter a subdomain that will be used to construct the URL for the web application. Make sure it contains URL friendly characters.
    • Description – Enter a description for the application.
    • Script – Select the path to the startup script.
    • Runtime
      • Editor – Workbench
      • Kernel – Select Cloudera Data Visualization for the kernel supported by the Runtime variant of the CML project.
      • Edition – Select the edition of the Runtime variant you want to use for your application.
  5. Click Create Application.

In a few minutes, you should see the application status change from Starting to Running on the Applications page. Your application has a hosted, fully-functional Data Visualization platform.

You can restart, stop, or delete a CDSW application from the supplemental menu of the application. If you want to make changes to the application, click Application Details and go to the Settings tab.

Start Data Visualization.