Introduction to Azure

Azure supports the following popular programming languages:

• Python
• JavaScript
• Java
• .NET
• Go

It also offers a comprehensive SDK library and toolkits for the following:

• VS Code
• Visual Studio
• IntelliJ
• Eclipse

Azure takes advantage of the existing developer skills and makes them productive right away. We can incorporate Azure into any application in different ways depending on our needs.

Application hosting on Azure

Azure can host our entire application stack, from web applications and APIs to databases and storage services. Azure supports various hosting models, from fully managed services to containers to virtual machines (VMs). Our applications can utilize the scalability, high availability, and security built into Azure when we’re using fully managed Azure services.

Key services

• Azure Machine Learning and artificial intelligence services empower developers and data scientists with a wide range of productive capabilities for building, training, and deploying machine learning (ML) models.

• Azure analytics services help gather, process, analyze, and visualize data of any volume.

• Azure allows us to host VMs on the cloud either from a virtual hard drive or an array of templates that Azure provides.

• Azure also provides cloud-based storage to store applications and backup data safely and securely. It supports faster innovation with secured and fully managed database services.

• Azure app services provide scalable hosting platforms. We can quickly deploy, operate, and scale entire apps with them.

AI and ML services offered by Azure

• Anomaly Detector: Adds anomaly detection capabilities to apps.

• Azure Cognitive Search: Adds enterprise-scale search capabilities to apps.

• Azure Machine Learning: Enterprise-grade ML capabilities from building to deploying ML models.

• Azure Cognitive Services: Pretrained models that can be consumed by invoking the REST API. They can be broadly classified into the following:

  • Language services
  • Vision services
  • Speech services
  • Decision services

• Azure Bot Services: Builds automatic conversation capabilities for customers, allowing them to customize per domain.

• Azure Applied AI Services: Specialized services that enable organizations to accelerate time for standard applications.

• Azure Open Datasets: Curated datasets hosted by the platform and used in building ML models.

• Data Science Virtual Machines (DSVMs): Pre-configured environment for AI development.

• Kinect DK: An SDKSoftware Development Kit bundled with computer vision and speech models that support advanced sensors.

• Project Bonsai: Creates intelligent control systems.

• Azure OpenAI Service: Applies advanced language models to various industry use cases.

• Azure Metrics Advisor: An AI monitor that tracks metrics and diagnoses issues.

• Microsoft Genomics: Powers genome sequencing.

Azure service classification

The services provided by Azure can be classified into on-premises support, infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (SaaS) models. We can see how Azure can manage the different service models in the example below. The cells marked with Y can be managed by Azure, whereas the cells marked with N can be managed by an individual.

On-premises support (serverless): In this role, the application can leverage the important pieces from Azure, like Azure Cognitive Services or Azure Storage Services. This would be a loose coupling with Azure and require minimal design changes in the existing application.

Infrastructure as a service (IaaS): Using infrastructure pieces offered by Azure, we can build innovative, scalable, and highly secured applications. For example, we can use Azure Virtual Machines to manage the servers, or we can use Azure Disk Storage to manage storage servers.

Platform as a service (PaaS): Platforms include caching, queues, and data storage to migrate to Azure. PaaS reduces the time and cost of managing servers, storage, networking, and other application infrastructure. We can use PaaS services like Azure App Service for hosting a web service and Azure Cognitive Search instead of Elasticsearch.

Software as a service (Saas): We can connect to SaaS products from Azure over the internet. Common examples of Azure SaaS services are email, calendaring, and office tools.

Azure interfaces

Azure provides multiple interfaces to manage services and resources. We will cover the important ones.

Azure Portal

Azure Portal is a web-based, unified console for managing Azure resources. With the Azure Portal, we can do the following:

• Manage subscriptions
• Manage and monitor the application information
• Create, delete, and manage Azure resources

Azure Marketplace

Azure Marketplace helps connect with Microsoft partners, vendors, and startups. Azure Marketplace customers can find, try, purchase, and provision applications and services from hundreds of leading software providers.

Now we understand the broad capabilities of Azure. Let’s dive into the ML, storage, and computing capabilities.