Chapter 4 of 21
Cloud Service Models: IaaS, PaaS, and SaaS in Azure
Move beyond buzzwords and see how Infrastructure as a Service, Platform as a Service, and Software as a Service shape real Azure solutions and exam scenarios.
Orienting: Where Service Models Fit in AZ-900
Where Service Models Fit
You already met cloud computing, deployment models, and the shared responsibility model. Now we focus on what you actually consume from Azure: cloud service models.
Canonical List (In Order)
For AZ-900 you must know the canonical service models, in order: 1) Infrastructure as a Service (IaaS), 2) Platform as a Service (PaaS), 3) Software as a Service (SaaS).
What You Need to Do
You must recall the canonical definitions, match Azure services and Microsoft products to each model, and compare them by control vs responsibility.
Intuition for the Models
IaaS = raw building blocks; PaaS = ready-made platform for apps; SaaS = finished app you just use. AZ-900 loves scenario questions that test this.
Canonical Definitions: IaaS, PaaS, SaaS
IaaS Canonical Definition
"Infrastructure as a Service (IaaS) is a cloud service model that provides virtualized computing resources such as servers, storage, and networking on demand."
PaaS Canonical Definition
"Platform as a Service (PaaS) is a cloud service model that provides a complete development and deployment environment in the cloud, including infrastructure, middleware, and development tools."
SaaS Canonical Definition
"Software as a Service (SaaS) is a cloud service model that delivers software applications over the internet on a subscription basis."
Canonical List in Order
The canonical list, in order, is: 1) Infrastructure as a Service (IaaS), 2) Platform as a Service (PaaS), 3) Software as a Service (SaaS).
Visualizing the Stack: On-Prem vs IaaS vs PaaS vs SaaS
The Application Stack
Picture layers: datacenter, hardware, virtualization, OS, middleware, application, data. Service models decide who manages which layers.
On-Prem vs IaaS
On-prem: you manage everything. In IaaS, Azure manages datacenters, hardware, virtualization; you manage OS, middleware, app, and data.
PaaS Responsibility
In PaaS, Azure manages up through middleware and runtime. You focus on application code and data, not OS patching or web server setup.
SaaS Responsibility
In SaaS, Azure or Microsoft manages the entire stack. You mainly configure the app and manage users and data within it.
Azure IaaS in Practice: Virtual Machines and More
IaaS Definition Reminder
"Infrastructure as a Service (IaaS) is a cloud service model that provides virtualized computing resources such as servers, storage, and networking on demand."
Core Azure IaaS Services
Key IaaS examples: Azure Virtual Machines, Azure Virtual Network, Azure Load Balancer, Azure Disk Storage, and Azure Blob Storage used as raw storage.
Lift-and-Shift Scenario
Legacy app on Windows Server moves to Azure using VMs, VNets, and disks. The company still patches Windows and manages IIS: this is classic IaaS.
Exam Clue for IaaS
If a question stresses full OS control or installing custom software at OS level, think Azure Virtual Machines and IaaS.
Azure PaaS in Practice: App Service, Functions, SQL Database
PaaS Definition Reminder
"Platform as a Service (PaaS) is a cloud service model that provides a complete development and deployment environment in the cloud, including infrastructure, middleware, and development tools."
Core Azure PaaS Services
Key PaaS examples: Azure App Service, Azure Functions, Azure SQL Database, and often Azure Container Apps or AKS in managed scenarios.
Modern Web App Scenario
A startup builds a web app using Azure App Service and Azure SQL Database. Azure manages OS and runtime; the team focuses on code and data: this is PaaS.
Exam Clue for PaaS
If the question stresses rapid development, no server management, and deploying code or databases, think PaaS services like App Service or Azure SQL Database.
SaaS in the Microsoft Cloud: Microsoft 365, Dynamics 365, and More
SaaS Definition Reminder
"Software as a Service (SaaS) is a cloud service model that delivers software applications over the internet on a subscription basis."
Microsoft SaaS Examples
Key SaaS offerings: Microsoft 365, Dynamics 365, Power BI service, plus many third-party SaaS apps integrated with Microsoft Entra ID.
Collaboration Scenario
A company adopts Microsoft 365 for email and Teams. They manage users and policies, but Microsoft runs the apps: this is SaaS.
Exam Clue for SaaS
If you see end users consuming apps (email, CRM, analytics) with no code deployment or server management, choose SaaS.
Control vs Responsibility: Comparing IaaS, PaaS, SaaS
Shared Responsibility Reminder
The shared responsibility model defines how security and compliance responsibilities are divided between the cloud provider and the customer.
IaaS Control
In IaaS, you control OS, middleware, apps, and data. Azure manages datacenters, hardware, and virtualization. High control, high responsibility.
PaaS Control
In PaaS, Azure manages OS and middleware. You focus on application code and data. Less operational work, still significant control.
SaaS Control
In SaaS, the provider runs the full stack. You mainly manage users, configuration, and data use. Lowest operational burden, least low-level control.
Thought Exercise: Classify These Azure Services
Test your intuition by classifying real Azure services into IaaS, PaaS, or SaaS. Think through each one before checking the guidance.
- Azure Virtual Machines
- Ask yourself: Do I manage the OS and installed software?
- If yes, that strongly suggests IaaS.
- Azure App Service (Web Apps)
- You deploy code, but you never log into the underlying Windows or Linux OS.
- Azure handles OS patching and web server configuration.
- That is PaaS.
- Azure SQL Database
- You manage databases and queries, but not the SQL Server OS or installation.
- This is a managed database platform: PaaS.
- Microsoft 365 Exchange Online
- You configure mailboxes and policies; Microsoft runs the mail servers.
- This is clearly SaaS.
- Azure Virtual Desktop
- You provide Windows desktop images and apps; Azure provides the infrastructure and control plane.
- In AZ-900, this is often treated as IaaS-based (infrastructure plus management), but the key is: you still manage the Windows environment itself.
- Power BI service (cloud workspace and dashboards)
- End users consume analytics via a web app.
- This is SaaS.
As you continue, try to justify each classification in terms of who manages OS, middleware, and the application.
Quick Check 1: Match Scenario to Service Model
Use this quiz to confirm you can match scenarios to IaaS, PaaS, and SaaS.
A development team wants to deploy a web API without managing the underlying Windows or Linux servers. They only want to push code and let Azure handle OS patching and web server configuration. Which cloud service model best fits this need?
- Infrastructure as a Service (IaaS)
- Platform as a Service (PaaS)
- Software as a Service (SaaS)
- Hybrid cloud
Show Answer
Answer: B) Platform as a Service (PaaS)
This is **Platform as a Service (PaaS)**. The team deploys code while Azure manages the OS and middleware. IaaS would require them to manage the OS; SaaS would be consuming a finished application; hybrid cloud is a deployment model, not a service model.
Quick Check 2: Canonical Order and Definitions
Now confirm you remember the canonical order and definitions.
Which option lists the cloud service models in the correct canonical order, from lowest-level building blocks to highest-level finished applications?
- Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure as a Service (IaaS)
- Infrastructure as a Service (IaaS), Software as a Service (SaaS), Platform as a Service (PaaS)
- Platform as a Service (PaaS), Infrastructure as a Service (IaaS), Software as a Service (SaaS)
- Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS)
Show Answer
Answer: D) Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS)
The canonical list, in order, is: **Infrastructure as a Service (IaaS)**, **Platform as a Service (PaaS)**, **Software as a Service (SaaS)**. Remember: IaaS = infrastructure, PaaS = platform, SaaS = software.
Flashcards: Lock In the Core Definitions
Use these flashcards to reinforce the exact wording and key examples for each service model.
- Infrastructure as a Service (IaaS) – canonical definition
- Infrastructure as a Service (IaaS) is a cloud service model that provides virtualized computing resources such as servers, storage, and networking on demand.
- Platform as a Service (PaaS) – canonical definition
- Platform as a Service (PaaS) is a cloud service model that provides a complete development and deployment environment in the cloud, including infrastructure, middleware, and development tools.
- Software as a Service (SaaS) – canonical definition
- Software as a Service (SaaS) is a cloud service model that delivers software applications over the internet on a subscription basis.
- Canonical list of cloud service models (in order)
- 1. Infrastructure as a Service (IaaS) 2. Platform as a Service (PaaS) 3. Software as a Service (SaaS)
- Azure examples of IaaS
- Typical IaaS in Azure: Azure Virtual Machines, Azure Virtual Network, Azure Load Balancer, Azure Disk Storage, Azure Blob Storage used as raw storage.
- Azure examples of PaaS
- Typical PaaS in Azure: Azure App Service, Azure Functions, Azure SQL Database, and often Azure Container Apps or AKS in managed scenarios.
- Microsoft examples of SaaS
- Typical SaaS: Microsoft 365 (Exchange Online, SharePoint Online, Teams), Dynamics 365, Power BI service, and many third-party SaaS apps integrated with Microsoft Entra ID.
- Shared responsibility model – definition
- The shared responsibility model is a framework that defines how security and compliance responsibilities are divided between the cloud provider and the customer.
- High-level: what you manage in IaaS vs PaaS vs SaaS
- IaaS: you manage OS, middleware, apps, data. PaaS: you manage apps and data. SaaS: you mainly manage users, configuration, and data usage.
Mini Design Challenge: Choose the Right Service Model
Apply everything you have learned to a short design exercise. For each scenario, pause and decide: IaaS, PaaS, or SaaS? Then compare with the guidance.
- Legacy finance system
- An organization has a complex legacy finance app that requires specific OS settings and custom drivers. Rewriting it is not feasible this year.
- Best fit: IaaS (Azure Virtual Machines). They need deep OS-level control.
- New mobile backend API
- A team is building a new mobile app and wants a REST API. They want automatic scaling and do not want to manage servers.
- Best fit: PaaS (Azure App Service or Azure Functions). They focus on code; Azure runs the platform.
- Company-wide CRM
- Sales and support teams need customer relationship management with minimal IT overhead.
- Best fit: SaaS (Dynamics 365). It is a finished CRM application.
- Analytics dashboards for executives
- Executives want dashboards and reports with minimal setup. Data engineers can prepare data, but they do not want to host BI servers.
- Best fit: SaaS (Power BI service) for the dashboards, possibly combined with PaaS data services behind the scenes.
- Exam reflection
- For each choice, ask yourself: Who manages the OS? Who deploys code? Who owns the application itself? Those answers almost always point to the correct service model in AZ-900 scenarios.
Key Terms
- Dynamics 365
- A Microsoft SaaS suite of intelligent business applications for CRM and ERP scenarios.
- Microsoft 365
- A Microsoft SaaS offering that includes cloud-based productivity services such as Exchange Online, SharePoint Online, Teams, and Office applications.
- Azure Functions
- An Azure PaaS/serverless compute service that runs event-driven code without requiring you to provision or manage servers.
- Power BI service
- A Microsoft SaaS analytics service that provides interactive visualizations and business intelligence capabilities via the cloud.
- Azure App Service
- An Azure PaaS service for hosting web apps and APIs where you deploy code and Azure manages the underlying infrastructure and runtime.
- Azure SQL Database
- A fully managed PaaS relational database service in Azure where Microsoft manages the database engine and infrastructure.
- cloud service models
- The canonical list of cloud service models, in order: 1) Infrastructure as a Service (IaaS), 2) Platform as a Service (PaaS), 3) Software as a Service (SaaS).
- Azure Virtual Machines
- An Azure IaaS service that provides virtualized Windows or Linux servers where you manage the operating system, applications, and data.
- shared responsibility model
- The shared responsibility model is a framework that defines how security and compliance responsibilities are divided between the cloud provider and the customer.
- Platform as a Service (PaaS)
- Platform as a Service (PaaS) is a cloud service model that provides a complete development and deployment environment in the cloud, including infrastructure, middleware, and development tools.
- Software as a Service (SaaS)
- Software as a Service (SaaS) is a cloud service model that delivers software applications over the internet on a subscription basis.
- Infrastructure as a Service (IaaS)
- Infrastructure as a Service (IaaS) is a cloud service model that provides virtualized computing resources such as servers, storage, and networking on demand.