Module 1: Azure Foundations
Cloud Concepts on Azure
Before you learn any Azure service, you need the vocabulary that every cloud conversation uses. Managers, auditors, developers and the AZ-900 exam all assume you can explain what the cloud is, why organisations move to it, and how responsibility is split between you and Microsoft. This lesson gives you that foundation, so that every later lesson has somewhere to attach.
What cloud computing actually means
Cloud computing is the delivery of computing services (servers, storage, databases, networking, software and analytics) over the internet, on demand, with you paying for what you use. Instead of buying a server, waiting weeks for delivery and installing it in a back office in Westlands, you request a virtual server in Azure and it is running within minutes. When you no longer need it, you delete it and the charges stop.
Microsoft describes several characteristics that make something "cloud" rather than ordinary hosting:
- On-demand self-service: you create resources yourself through a portal, command line or code, without phoning a supplier.
- Broad network access: services are reachable over the network from many kinds of device.
- Resource pooling: Microsoft runs huge shared datacentres and allocates capacity to many customers.
- Rapid elasticity: you can add or remove capacity quickly as demand changes.
- Measured service: usage is metered, so you are billed for what you consume.
CapEx, OpEx and the consumption model
Traditional IT is mostly capital expenditure (CapEx): you spend a large amount up front on hardware that you then own and depreciate over several years. Cloud computing is mostly operational expenditure (OpEx): you pay regularly for services as you use them, with little or no up-front cost. This is called the consumption-based model.
The consumption model has practical benefits. There is no need to guess future capacity and buy for the peak. You can try an idea cheaply and stop if it fails. The trade-off is that costs are variable, so you must monitor them, which you will learn to do in the governance module.
Benefits you will be asked about
The AZ-900 study guide expects you to describe the benefits of cloud services. Learn these terms precisely:
| Benefit | Meaning |
|---|---|
| High availability | The service stays accessible with minimal downtime, usually backed by a service-level agreement (SLA). |
| Scalability | Adding resources to handle more load. Vertical scaling means a bigger machine; horizontal scaling means more machines. |
| Elasticity | Scaling automatically up and down as demand changes. |
| Reliability | The ability to recover from failures and keep working, helped by Azure's global spread of datacentres. |
| Predictability | Predictable performance and predictable cost, supported by tools such as autoscaling and cost calculators. |
| Security and governance | Built-in tools for policies, compliance and protection, plus the option to patch automatically on managed services. |
| Manageability | Management of the cloud (scaling, monitoring) and management in the cloud (portal, CLI, APIs). |
Service models and the shared responsibility model
Cloud services come in three main models:
- Infrastructure as a Service (IaaS): you rent virtual machines, storage and networks. You manage the operating system, patches and applications. Example: Azure Virtual Machines.
- Platform as a Service (PaaS): Microsoft manages the operating system and runtime; you deploy your code and data. Example: Azure App Service, Azure SQL Database.
- Software as a Service (SaaS): you use finished software over the internet. Example: Microsoft 365.
The shared responsibility model explains who secures what. Microsoft is always responsible for the physical datacentre, physical network and physical hosts. You are always responsible for your data, your devices, and your accounts and identities. Everything in between (operating system, network controls, applications) shifts from you towards Microsoft as you move from IaaS to PaaS to SaaS.
There are also three deployment models: public cloud (Azure, shared infrastructure owned by Microsoft), private cloud (cloud-style infrastructure used by one organisation, often in its own datacentre) and hybrid cloud (a combination, connected together). Azure Arc is the Azure service that helps you manage resources outside Azure, such as on-premises servers, from Azure.
Worked example: a Nairobi retailer
Imagine a clothing retailer in Nairobi with an online shop. Sales are steady most of the year but spike sharply in December. On-premises, they would have to buy servers big enough for December and leave them mostly idle for eleven months. In Azure they can run the shop on App Service (PaaS), scale out to more instances in December and back in January, and pay for that extra capacity only while they use it. Because App Service is PaaS, Microsoft patches the operating system; the retailer remains responsible for its code, its customer data and who has access to the account.
Common mistakes to avoid
- Thinking the cloud is automatically cheaper. It is cheaper only when you size resources sensibly and switch off what you do not use.
- Believing Microsoft is responsible for your data. Under every service model, your data and identities remain your responsibility.
- Mixing up scalability and elasticity. Elasticity is scalability that happens automatically in response to demand.
- Assuming hybrid means "half migrated". Hybrid is a deliberate model that many organisations keep long term.
Exercise
Pick a business you know, such as a pharmacy, school or SACCO. Write down three IT workloads it runs (for example email, accounting, a website) and classify each as a candidate for IaaS, PaaS or SaaS, with one sentence explaining why. Then, for one workload, list two things Microsoft would be responsible for and two things the business would remain responsible for.
Key points
- Cloud computing means on-demand, metered computing services delivered over the internet.
- The cloud shifts spending from CapEx to OpEx through a consumption-based model.
- IaaS, PaaS and SaaS move more responsibility to Microsoft, but your data and identities always stay with you.
- Know the exact meaning of high availability, scalability, elasticity, reliability and predictability.