Azure Machine Learning - Create Compute Instance And Compute Cluster

In this article, we’ll learn about Azure Machine Learning and create computer cluster and compute instance in Machine Learning Workspace in Azure which we’ll use for our project on the Azure Machine Learning Series. This article is a part of the Azure Machine Learning Series where we’ll learn about the end-to-end process of Machine Learning capabilities enabled by Azure Machine Learning Studio.

  1. Azure Machine Learning - Create Workspace for Machine Learning
  2. Azure Machine Learning – Create Compute Instance and Compute Cluster
  3. Azure Machine Learning - Writing Python Script in Notebook
  4. Azure Machine Learning - Model Training

Microsoft AI

Microsoft AI is a powerful framework that enables organizations, researchers, and non-profits to use AI technologies with its powerful framework which offers services and features across domains of Machine Learning, Robotics, Data Science, IoT, and many more.  Learn more about Microsoft AI from this article.

Azure Machine Learning

The Azure Machine Learning enriches and consolidates the functionalities to support model training and deployment which transitions from Machine Learning Studio. It provides tools for Machine Learning works for all skill levels, provides an open and interoperable framework with support to different languages, and enables robust end-to-end MLOps. It also supports Automated Machine Learning. Read this article Auto ML to learn more about it.

So, where and how do we start if we want to create and deploy a Machine Learning project? Azure Machine Learning provides all the tools through its portal to create the resources and set up the infrastructure that is needed for any kind of machine learning works.


Before we start with the step-by-step guide to create the compute instance and compute cluster, it is essential that the Machine Learning Workspace has been created. Follow the article, Azure Machine Learning - Create Workspace for Machine Learning, to create the workspace.

Once, you have successfully created the Machine Learning Workspace, you can follow up the following steps.

Step 1

Visit the Azure Workspace you’ve created.

You can visit the Azure Machine Learning Studio from here, where you’ll see the workspace name you had created.

The Azure ML Studio home page will look similar to as follows.

Create Compute Instance

Step 2

Next, Click on Compute under Manage.

Here, Select the Compute Instance. Under this section, Select +New Button.

Step 3

You’ll be showcased with the settings to be filled with. Just go with the default selection as of now.

Fill in the Compute Name of you choice. Under the Virtual machine type, have the CPU selected. For the virtual machine size, Standard_DS3_v2 is good to go for general purpose. As we are going to learn about Model Training and more in the following articles, it is better to have it selected over other options.

Step 4

You can explore the Advanced Settings but setting all as default is good for us as we do not require any special functionality.

Now, Click on Create.

Step 5

The Compute Instance creation process will begin which will take 2-3 mins usually. You can see the update under the State.

Step 6

Once, the provisioning is done and we’ll be notified.

Now, if we Refresh the page, we can see, the update on the State as Running.

Our Compute Instance is running now. You can select it and find its details.

Create Compute Clusters

Step 7

For Compute Clusters, similarly to Compute Instances, Select the Compte Clusters under Compute.

Step 8

Click on the +New button.

Step 9

Use the default selection. From the recommended option, select the Standard_DS11_v2. Click on Next.

Step 10

Now, name your Compute Cluster.

Under Minimum number of nodes, set it to 0 and for Maximum number of nodes, you can go from 1 to 4 as you choose. Also set the idle seconds before scale down between 120 seconds to 2400 as you please..

Now, Click on Create.

Step 11

We’ll be notified with the success of the provisioning of the compute cluster.

As we refresh, we can see the change in State as Succeeded from the Creating and Resizing of earlier.

With this, we have now created both compute instance and compute clusters.


Step 12

As we have created the resources, later when we do not require it, it is essential that we delete them. One of the ways is selecting the Instance or Clusters as shown below and then Clicking on delete and confirming it.

Another way is to visit the Azure Portal and Delete the entire resource group itself. This will delete other numerous resources created while creating the ML Workspace from the workspace itself to Key Vault, Storage, Application Insight, Container Registry, and more.


Thus, in this article, we’ve learned to create Compute Instance and Compute Cluster. This will be an essential step with the learnings we’ll do forward in this series from using the Notebook to Training our Machine Learning Model.

Similar Articles