This chapter introduces cloud computing itself — what it means to use computing resources delivered over the internet instead of running your own hardware. AZ-900 tests this under objective 1.1, "Describe cloud concepts," the foundation the rest of the exam builds on.
Jump to a section
A simple way to picture What is Cloud Computing?
Owning on-premises servers is like building your own house: you buy the land, pour the foundation, install plumbing and electrical, and you're responsible for every repair forever. Cloud computing is like renting a furnished apartment instead — the building, utilities, and maintenance are already handled by someone else, and you pay for what you use, month to month, only for the space you actually occupy. You can move to a bigger apartment next month if your needs grow, or a smaller one if they shrink, without the years-long commitment of building and owning a house. That flexibility — using computing resources without owning or maintaining the underlying infrastructure yourself — is the core idea behind cloud computing.
What cloud computing is
Cloud computing is the delivery of computing resources — servers, storage, databases, networking, software — over the internet, provided and maintained by a cloud provider like Microsoft Azure, rather than owned and run by the customer on their own physical hardware.
How it works
Instead of buying physical servers and installing them in an on-premises data center, a customer requests resources (a virtual machine, a database, storage space) from Azure, and Azure provisions them from its own large-scale infrastructure. The customer accesses and manages those resources remotely — through the Azure portal, command-line tools, or APIs — without needing to touch or maintain the physical hardware underneath.
Key characteristics
On-demand self-service: resources can be provisioned without needing to call a vendor or wait for physical hardware to arrive.
Broad network access: resources are reachable over the internet from a wide range of devices.
Resource pooling: the provider's physical infrastructure serves many customers, with resources allocated dynamically as needed.
Rapid elasticity: capacity can scale up or down relatively quickly as demand changes.
Measured/metered usage: customers are typically billed based on what they actually consume.
How it compares to the on-premises equivalent
Running infrastructure on-premises means the organization owns the hardware outright, handles all maintenance and eventual replacement, and typically has to plan and pay for peak capacity in advance, even if that capacity sits unused most of the time. Cloud computing shifts that responsibility and much of that upfront capital cost to the provider, in exchange for ongoing usage-based payments.
Azure portal and CLI touchpoints
New Azure resources are typically created through the Azure portal (a web-based interface), the Azure CLI, or Azure PowerShell:
az vm create --resource-group myRG --name myVM --image Ubuntu2204All three ultimately go through Azure Resource Manager, covered in a later chapter.
Recognize the shift from ownership to consumption
The core mental shift cloud computing represents is moving from owning fixed infrastructure to consuming resources as a service, paying based on usage rather than upfront hardware purchases.
Identify the resource being requested
Whether it's compute (a VM), storage, or a database, the customer specifies what they need, and Azure provisions it from its underlying infrastructure without the customer managing physical hardware.
Access and manage remotely
Once provisioned, resources are managed through the portal, CLI, PowerShell, or APIs — not by physically visiting a data center.
Pay based on consumption
Billing is typically tied to actual usage (time running, data stored, requests made) rather than a fixed upfront cost, though the exact pricing structure is covered in a dedicated pricing chapter.
A startup building a new web application uses Azure virtual machines and a managed database instead of purchasing and racking physical servers — letting a small team launch quickly without hiring dedicated data-center staff. As the application grows, they can provision additional resources rather than waiting weeks for new hardware to be delivered and installed.
A larger enterprise running seasonal workloads (like a retailer during a holiday sales period) uses cloud resources to scale up capacity temporarily for the busy period, then scale back down afterward — something that would be far less practical with owned, on-premises hardware sized for peak demand year-round. Organizations that don't think through this consumption model sometimes provision cloud resources they no longer need and forget to scale back down, leading to unexpectedly high bills — a common real-world pitfall.
Objective 1.1 expects a clear understanding of what cloud computing is and its key characteristics — on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured usage.
A common wrong answer is describing cloud computing as simply "using the internet" — the exam expects the more specific concept of consuming computing resources (compute, storage, networking, etc.) delivered by a provider, not just general internet usage. Another trap is confusing cloud computing with any single deployment or service model (those are covered separately) rather than understanding it as the broader concept those models sit under.
Stable terms: on-demand self-service, resource pooling, elasticity, measured service. These are the widely recognized characteristics of cloud computing and are safe to know by name.
Memory trick: if a question describes paying for what you use, accessing resources remotely without owning hardware, and scaling up or down as needed — that's describing cloud computing itself, the foundational concept underneath everything else on this exam.
Cloud computing is the delivery of computing resources over the internet by a provider, rather than the customer owning and maintaining physical hardware.
Key characteristics include on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured/metered usage.
Resources are provisioned and managed remotely — through the portal, CLI, PowerShell, or APIs.
Cloud computing shifts responsibility for physical infrastructure to the provider, in exchange for usage-based payment.
These come up on the exam all the time. Here's how to tell them apart.
On-Premises Infrastructure
Organization owns and maintains physical hardware
Capacity must be purchased in advance for peak demand
Scaling requires buying and installing new hardware
Large upfront capital investment
Cloud Computing
Provider owns and maintains physical hardware
Capacity can be adjusted closer to actual demand
Scaling is a configuration change, not a hardware purchase
Ongoing usage-based operating expense
Mistake
Cloud computing just means 'using the internet.'
Correct
It specifically means consuming computing resources — compute, storage, networking, databases — delivered by a provider over the internet, not simply browsing the web or using any online service.
Mistake
Cloud computing always means lower cost than owning hardware.
Correct
It shifts cost from upfront capital expense to ongoing usage-based spending, which is often more efficient, but isn't automatically cheaper in every scenario — it depends on the workload and how well it's managed.
Mistake
Once something is 'in the cloud,' scaling and maintenance are entirely automatic with no customer involvement.
Correct
The provider maintains the underlying infrastructure, but the customer is still responsible for configuring, managing, and often scaling their own resources appropriately.
It's the delivery of computing resources — servers, storage, databases, networking, and more — over the internet by a provider, so customers can use them without owning or maintaining the physical hardware themselves. Azure, and providers like it, host the infrastructure and let customers provision and manage resources remotely.
No. The internet is the network that makes cloud computing accessible, but cloud computing specifically refers to consuming computing resources (compute, storage, etc.) as a service, not just any internet-based activity.
Not automatically — it shifts costs from large upfront purchases to ongoing usage-based spending, which is often more efficient, but actual savings depend on how well resources are managed. Poorly managed cloud usage can end up costly too.
You've just covered What is Cloud Computing? — now see how well it sticks with free AZ-900 practice questions. Full explanations included, no account needed.
Done with this chapter?