Describe Cloud Concepts · Topic 2 of 3
Benefits of Cloud Services
AZ-900 benefits of the cloud: high availability, scalability, elasticity, reliability, predictability, security, governance and manageability explained.
- 5 min
- High availability means a service stays up. It is measured by an SLA percentage.
- Vertical scaling changes the size of one machine. Horizontal scaling changes how many machines.
- Elasticity is automatic scaling to match demand.
What it is
Organizations move to the cloud because it gives them things that are hard or expensive to build themselves. The exam groups these benefits into a short list of terms.
You need to recognise each term from a one line description. This page defines them and shows how they differ.
How it works
High availability
A highly available service keeps running with little or no downtime. Providers achieve this with redundant hardware and by spreading workloads across locations.
Availability is promised in a service level agreement (SLA), usually as a percentage of uptime. A higher percentage means less allowed downtime.
Scalability and elasticity
Scalability means you can increase or decrease the resources a workload uses. There are two directions.
| Type | What changes | Example |
|---|---|---|
| Vertical scaling (scale up or down) | The power of one machine, such as more CPU or memory | Moving a virtual machine to a larger size |
| Horizontal scaling (scale out or in) | The number of machines or instances | Adding more web servers as traffic grows |
Elasticity is scaling that happens automatically as demand changes, so you add capacity at a busy time and release it afterward.
Reliability
Reliability is the ability of a system to recover from failures and keep working. The cloud supports it through a global network of datacenters, so a workload can run in more than one region or zone and survive a local failure.
Predictability
Predictability has two sides.
- Performance predictability: a workload behaves consistently under load, helped by autoscaling and load balancing.
- Cost predictability: you can forecast and track spending, using tools such as cost analysis and the pricing calculator.
Security and governance
Cloud providers invest heavily in physical security and offer security tools and certifications for compliance. You can also apply governance, which means rules for how resources are created and used, such as limits on regions or sizes.
Security in the cloud is still shared. The provider secures the platform, and you secure what you put on it.
Manageability
Manageability covers two ideas.
- Management of the cloud: automatic scaling, deploying resources from templates and monitoring that alerts you to problems.
- Management in the cloud: managing resources through a web portal, command line tools, scripts or an API.
Recognise the benefit from the scenario
Most questions describe a situation and ask which benefit it shows. These phrases point to the answer.
| The scenario says | The benefit |
|---|---|
| The website keeps running when a datacenter fails | High availability |
| More servers are added automatically on Black Friday | Elasticity, which is a kind of scalability |
| The application recovers by itself after a failure | Reliability |
| The monthly bill can be forecast from usage and tags | Cost predictability |
| Response time stays steady under heavy load | Performance predictability |
| Servers are deployed from a template in minutes | Manageability of the cloud |
| Resources are managed from the portal or scripts | Manageability in the cloud |
What an availability percentage means
A service level agreement (SLA) promises a percentage of uptime. The percentage looks small, but each extra nine removes a lot of allowed downtime.
| Availability | Allowed downtime per year |
|---|---|
| 99 percent | About 3.7 days |
| 99.9 percent | About 8.8 hours |
| 99.95 percent | About 4.4 hours |
| 99.99 percent | About 53 minutes |
| 99.999 percent | About 5 minutes |
If a service does not meet its SLA, the provider usually gives a service credit. This is a refund against future charges, not compensation for lost business.
Governance in the cloud
Governance means rules that keep resources consistent and compliant. In Azure these rules are enforced with tools such as Azure Policy, which blocks resources that break a standard, and resource locks, which stop accidental deletion. You meet both in the management and governance domain.
How the exam asks about this
- Two benefits often appear together as options. Choose the one that matches the verb: "stays up" is availability, "recovers" is reliability, "adds capacity" is scalability.
- "Which type of scaling adds instances?" Horizontal, also called scaling out.
- "Which benefit lets a company forecast spending?" Cost predictability.
- A question about an SLA percentage usually asks which number allows the least downtime. The highest percentage wins.
What to remember for the exam
- High availability means a service stays up. It is measured by an SLA percentage.
- Vertical scaling changes the size of one machine. Horizontal scaling changes how many machines.
- Elasticity is automatic scaling to match demand.
- Reliability means recovering from failure, often by using several regions or zones.
- Predictability covers both consistent performance and forecastable cost.
- Governance is about rules for resources. Manageability is about managing and automating them.
"Add more instances" points to horizontal scaling. "Move to a bigger VM" points to vertical scaling. "Scales automatically" points to elasticity.
Common mistakes
Mixing up scale up and scale out. Scale up makes one machine larger. Scale out adds more machines.
Treating scalability and elasticity as the same. Scalability is the ability to change resources. Elasticity is doing it automatically as demand moves.
Confusing availability with reliability. Availability is how much of the time a service is up. Reliability is how well it recovers from failures.
Assuming predictability only means cost. It also covers consistent performance.
A higher SLA percentage does not mean the service never fails. It means the provider commits to less allowed downtime.
Check yourself
Which scaling type adds more instances to handle load?
Horizontal scaling, also called scaling out.
What is elasticity?
Automatically adding or removing resources to match demand.
Which benefit describes recovering from a failure by running in more than one location?
Reliability.
Topic 2 of 3 in Describe Cloud Concepts
Test yourself on Describe Cloud Concepts
The full Microsoft AZ-900 question bank has 500 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by Microsoft. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.