Module 1 — Cloud foundations
Place in the route
Start here. This module gives the learner the vocabulary needed to reason about Azure services before learning their product names.
Outcomes
After this module, you should be able to explain cloud computing, identify the public/private/hybrid models, place IaaS, PaaS, and software as a service (SaaS) on the responsibility spectrum, and distinguish scalability, elasticity, availability, and reliability.
The cloud model
Cloud computing means consuming computing capability over a network instead of owning all of the physical equipment that supplies it. The provider operates the physical datacenter; the customer still owns decisions about data, identities, access, and the workload’s configuration. This is the shared-responsibility boundary, and it moves as the service becomes more managed.
Public cloud uses a provider’s shared infrastructure through isolated customer accounts. Private cloud dedicates cloud-like capabilities to one organization. Hybrid cloud connects environments so a workload or its data crosses that boundary. The model is an ownership and placement decision, not a synonym for a particular Azure service.
Service models and the decision boundary
IaaS supplies infrastructure such as virtual machines; the customer manages the guest operating system and application. PaaS supplies a managed runtime such as App Service; the customer manages application code and data while Azure operates more of the platform. SaaS supplies a finished application; the customer configures users, data, and use of the application. Serverless is a PaaS-style operating model in which the platform allocates execution capacity around events or requests rather than asking the customer to manage servers.
Choose the most managed model that still satisfies the workload’s control, runtime, portability, compliance, and integration needs. A VM is useful when an application needs OS-level control; App Service is useful when the team wants to deploy web code without patching a server; SaaS is useful when the organization needs a capability rather than ownership of its implementation. More control also means more patching, configuration, and failure responsibility.
Capacity, availability, and reliability
Scalability is the ability to handle more load by adding capacity. Vertical scaling changes the size of one instance; horizontal scaling adds instances. Elasticity is the ability to adjust capacity as demand changes. Availability describes whether a service is usable at a given time; reliability describes its ability to continue operating and recover predictably over time. A larger VM may scale vertically, but it does not by itself make a single-instance application highly available.
See cloud-deployment-models for the canonical deployment-model comparison.
Conceptual checkpoint
For a public customer website, explain why a hybrid connection might be needed, who patches the OS under IaaS and PaaS, and whether adding a second instance improves scalability, availability, or both. State the condition that would make each answer change.
Hands-on / what to try
Use a paper or local diagram to assign responsibility for identity, data, OS, runtime, and physical infrastructure across VM, App Service, and SaaS. No Azure resource is required.
Sources and re-audit
- AZ-900 course, checked 2026-08-31.
- AZ-900 study guide, skills measured 2026-07-20, checked 2026-08-31.
- Shared responsibility in the cloud, checked 2026-08-31.
Re-audit when service-model responsibility guidance or the AZ-900 source baseline changes.