While creating a machine learning workspace, several associated resources are also created by Azure, which make your work with the workspace more convenient.
Which of the following items is not an associated resource and needs to be created manually when needed?
Click on the arrows to vote for the correct answer
A. B. C. D.Answer: B.
Option A is incorrect because Azure Container Registry with docker containers is at your disposal for deploying your model, without any manual provisioning.
Option B is CORRECT because as the default (automatically created) storage for a workspace is an Azure Storage Account.
If you want to use Data Lake Storage Gen2, for example because of its capability to manage hierarchical namespaces, you need to create one manually.
Option C is incorrect because an Azure Storage Account is automatically created on creating the workspace.
It is the default datastore for the workspace.
It is a flat, non-hierarchical storage.
Option D is incorrect because an Application Insights is also created on creating a workspace in order to store monitoring data related to your models.
Reference:
When creating a machine learning workspace on Azure, there are several associated resources that are created automatically to make working with the workspace more convenient. These resources can include storage accounts, virtual networks, and application insights, among others. However, one of the items listed in the question is not an associated resource and needs to be created manually when needed.
Option A: Azure Container Registry Azure Container Registry is a service used to store and manage container images for Docker and other container platforms. This service is commonly used in conjunction with Azure Kubernetes Service (AKS) to deploy containerized applications. In the context of a machine learning workspace, container images can be used to package machine learning models and associated code for deployment. Azure Container Registry is an associated resource that is created automatically when creating a machine learning workspace, and therefore not the correct answer.
Option B: Azure Data Lake Storage Azure Data Lake Storage is a scalable and secure data lake that can be used to store and analyze big data. It supports a variety of analytics frameworks and provides granular access control to data. In the context of a machine learning workspace, Azure Data Lake Storage can be used to store large datasets used for training and testing machine learning models. Azure Data Lake Storage is an associated resource that is created automatically when creating a machine learning workspace, and therefore not the correct answer.
Option C: Azure Storage Account Azure Storage Account is a service used to store and manage unstructured data such as blobs, files, queues, and tables. In the context of a machine learning workspace, Azure Storage Account can be used to store datasets, model checkpoints, and other artifacts used in the machine learning workflow. Azure Storage Account is an associated resource that is created automatically when creating a machine learning workspace, and therefore not the correct answer.
Option D: Azure Application Insights Azure Application Insights is a service used to monitor the performance and usage of web applications. It provides real-time insights into application performance, user behavior, and usage trends. In the context of a machine learning workspace, Azure Application Insights can be used to monitor the performance of machine learning models in production. However, Azure Application Insights is not an associated resource that is created automatically when creating a machine learning workspace, and therefore the correct answer to the question. If you need to use Azure Application Insights, you will need to create it manually.