Challenge 05: PaaS — Platform as a Service

Estimated Time

20-30 min | Cost: Free | Domain: Cloud Concepts (25-30%)

Exam skills covered

• Describe Platform as a Service (PaaS)
• Identify appropriate use cases for PaaS
• Describe the shared responsibility model as it relates to PaaS

Overview

Platform as a Service (PaaS) is a middle ground between IaaS and SaaS. The cloud provider manages the infrastructure AND the platform (OS, runtime, middleware), while you focus only on your application and data.

Think of PaaS like a food truck rental: you get the truck fully equipped with a kitchen (platform), and you just bring your recipes and ingredients (application and data). You don't worry about the engine, tires, or electricity.

PaaS is ideal for developers who want to build applications without worrying about managing servers, patching operating systems, or configuring infrastructure.

Explore

Task 1: Understand PaaS responsibilities

Layer	Who manages it?
Data & access	You
Applications	You
Runtime	Azure
Operating system	Azure
Virtual machine	Azure
Network controls	Shared
Physical infrastructure	Azure

Compare this with IaaS — notice how much more Azure manages for you!

Task 2: Explore Azure PaaS services

1. In the Azure Portal, click Create a resource
2. Search for App Service (Web App) — the flagship PaaS service
3. Explore the creation form:
   • Notice you choose a runtime (Node.js, Python, .NET, Java) — not an OS
   • You don't configure VM size — you choose a plan (pricing tier)
   • No disk management, no OS patching
4. Click Cancel — don't create anything

Task 3: Compare popular PaaS services

Azure PaaS Service	What it does	You manage
Azure App Service	Host web apps and APIs	Your code + configuration
Azure SQL Database	Managed relational database	Queries + data
Azure Cosmos DB	Managed NoSQL database	Data + access policies
Azure Functions	Run code without servers	Function code only
Azure Kubernetes Service	Managed container orchestration	Container images + config

Task 4: PaaS vs IaaS comparison

Aspect	IaaS (VM)	PaaS (App Service)
OS patches	You apply them	Azure handles it
Scaling	You configure scale sets	Built-in auto-scale
Deployment	Install software on VM	Deploy your code/container
Cost model	Pay for VM uptime	Pay for plan + consumption
Time to deploy	Hours (setup OS, install runtime)	Minutes (push code)
Control	Full OS access	Application-level only

Task 5: When to use PaaS

PaaS is best when:

• Building new web applications or APIs
• You want to focus on code, not infrastructure
• You need managed databases without DBA overhead
• You want automatic scaling and high availability
• Rapid development and deployment is critical

PaaS is NOT ideal when:

• You need full OS-level control
• You have legacy apps that require specific OS configurations
• You need to install custom drivers or kernel modules

Azure CLI Alternative

# List available App Service plans (does not create anything)
az appservice list-locations --sku F1 --output table

# List available runtimes for App Service
az webapp list-runtimes --os linux --output table

Key Concepts

Concept	Description
PaaS	Cloud manages infrastructure + platform; you deploy applications
Azure App Service	PaaS for hosting web apps, APIs, and mobile backends
Managed database	Database service where Azure handles backups, patching, HA
Focus on code	PaaS lets developers concentrate on business logic
Less control, less responsibility	Trade-off: simpler management but less OS-level access

Knowledge Check

1. In the PaaS model, who is responsible for managing the operating system?

2. A developer wants to deploy a Python web application without managing servers or OS patches. Which service model is most appropriate?

3. Which of the following is an example of PaaS in Azure?

4. What is a disadvantage of PaaS compared to IaaS?

5. Azure SQL Database is an example of which cloud service model?

Learn More

• 📚 Study Guide AZ-900 — Curated study materials
• Microsoft Learn: Describe cloud service types
• What is PaaS?