AZ-900 Describe cloud concepts Practice Test 14 — 15 Questions

Question 1

According to Microsoft, which factor does NOT affect the SLA percentage for an Azure service?

Accepted Answer

The number of users concurrently accessing the service. The SLA percentage for an Azure service is determined by factors such as redundancy, fault tolerance, and resource configuration, not by the number of concurrent users. Microsoft's SLAs are based on uptime guarantees tied to architectural choices like availability zones, disk types, and availability sets, while user load is managed through scaling and is not a factor in the SLA calculation.

Answer

Whether resources are deployed across availability zones

Answer

Whether Premium SSD or Standard HDD disks are used for VMs

Answer

Whether a single VM or availability set is used

Question 2

Which of the following best describes Platform as a Service (PaaS)?

Accepted Answer

The provider manages hardware and OS while customers manage their applications and data. Platform as a Service (PaaS) provides a managed hosting environment where the cloud provider handles the underlying infrastructure, including hardware, virtualization, operating system, and middleware. The customer is responsible only for deploying and managing their applications and data, making option B correct because it accurately describes this division of responsibility.

Answer

The customer manages everything including hardware and networking

Answer

The customer accesses a complete application managed entirely by the provider

Answer

The provider manages physical servers while customers manage OS and applications

Question 3

What is disaster recovery in the context of cloud computing?

Accepted Answer

Restoring systems and data after a catastrophic event with minimal downtime. Disaster recovery (DR) in cloud computing focuses on restoring systems, applications, and data after a catastrophic event (e.g., natural disaster, cyberattack) to meet Recovery Time Objective (RTO) and Recovery Point Objective (RPO) targets. Azure Site Recovery orchestrates replication, failover, and failback of workloads between primary and secondary regions, ensuring minimal downtime and data loss. This is distinct from high availability, which prevents outages, or auto-scaling, which handles demand fluctuations.

Answer

Preventing any service outages from occurring

Answer

Automatically scaling resources when demand increases

Answer

Distributing traffic across multiple servers for performance

Question 4

What does 'geo-distribution' mean in cloud computing?

Accepted Answer

Deploying applications and data across multiple geographic locations worldwide. Geo-distribution in cloud computing refers to deploying applications, data, and services across multiple geographically separated data centers or regions. This ensures low latency for users worldwide, improves availability through redundancy, and supports disaster recovery by isolating failures to a single region. Azure implements this through paired regions and traffic routing policies like performance-based routing in Azure Traffic Manager.

Answer

Distributing compute resources across multiple virtual machines in a single data center

Answer

Routing network traffic to the fastest available server

Answer

Encrypting data before sending it over the network

Question 5

What is the key difference between capital expenditure (CapEx) and operational expenditure (OpEx) in the context of cloud computing?

Accepted Answer

CapEx is upfront investment in owned infrastructure; OpEx is ongoing pay-as-you-go service costs. Capital expenditure (CapEx) involves a large upfront investment to purchase and own physical infrastructure (servers, storage, networking), while operational expenditure (OpEx) represents ongoing, consumption-based costs where you pay only for the resources you use (e.g., per-hour VM billing, per-GB storage fees). In cloud computing, the shift from CapEx to OpEx is a fundamental financial model change, enabling organizations to avoid large capital outlays and instead align costs with actual usage.

Answer

CapEx is for cloud spending; OpEx is for on-premises spending

Answer

CapEx and OpEx are identical in cloud environments

Answer

OpEx covers hardware costs; CapEx covers software licensing costs

Question 6

What is the difference between vertical scaling and horizontal scaling?

Accepted Answer

Vertical scaling increases a single resource's capacity; horizontal scaling adds more instances. Vertical scaling (scaling up) increases the capacity of a single resource, such as adding more RAM, CPU, or disk space to a virtual machine. Horizontal scaling (scaling out) adds more instances of a resource, distributing the load across multiple machines. This distinction is fundamental in cloud computing, where Azure Virtual Machine Scale Sets exemplify horizontal scaling by automatically adding VM instances, while resizing a VM to a larger SKU represents vertical scaling.

Answer

Vertical scaling adds more instances; horizontal scaling increases instance size

Answer

Both vertical and horizontal scaling achieve the same result through different means

Answer

Vertical scaling is for storage; horizontal scaling is for compute

Question 7

What is 'fault tolerance' in cloud computing?

Accepted Answer

The ability to continue operating correctly despite component failures. Fault tolerance in cloud computing refers to a system's ability to continue operating correctly, without interruption, even when one or more of its components fail. This is achieved through redundancy—such as duplicate hardware, software, or data paths—so that if a component fails, another automatically takes over without any impact on the user. It is a core design principle for high-availability systems, ensuring zero downtime despite failures.

Answer

The ability to automatically scale resources during peak usage

Answer

The ability to restore data after a major disaster

Answer

The ability to deploy applications in multiple geographic regions

Question 8

Which cloud concept allows businesses to deploy applications to global markets quickly without building physical infrastructure in those locations?

Accepted Answer

Global reach through cloud infrastructure. Global reach through cloud infrastructure enables businesses to deploy applications in multiple geographic regions using the cloud provider's existing data centers, eliminating the need to build and manage physical infrastructure in each target market. Providers like Azure offer region pairs and geographies that allow low-latency access and compliance with local data residency requirements, enabling rapid global expansion.

Answer

Elasticity

Answer

Capital expense avoidance

Answer

Fault tolerance

Question 9

Which of the following is NOT a benefit of moving to the cloud?

Accepted Answer

Elimination of internet connectivity requirements. Moving to the cloud does not eliminate the need for internet connectivity; in fact, cloud services are accessed over the internet or dedicated network connections. Option B is correct because it presents a false benefit—cloud computing still requires network connectivity to reach the provider's endpoints, and without it, resources become inaccessible. The other options are legitimate benefits: faster provisioning, global data center reach, and converting CapEx to OpEx.

Answer

Reduced time to provision new resources

Answer

Access to a global network of data centers

Answer

Conversion of capital expenses to operational expenses

Question 10

Which statement correctly describes the operational cost model of cloud computing?

Accepted Answer

You pay for cloud resources only when you consume them, on a recurring basis. Cloud computing follows an operational expenditure (OpEx) model where you pay only for the resources you actually consume, such as compute hours, storage GB-months, or data transfer, on a recurring basis. This eliminates the need for large upfront capital investments and allows you to scale costs with usage, aligning with the pay-as-you-go pricing model central to Azure and other cloud providers.

Answer

You make large upfront investments in hardware that you own and depreciate

Answer

You purchase cloud capacity in bulk at the beginning of each year

Answer

You lease physical servers from the cloud provider for a fixed monthly fee

Question 11

What is Software as a Service (SaaS)?

Accepted Answer

A model where the provider delivers complete software applications managed entirely by the provider. Software as a Service (SaaS) is a cloud computing model where the provider hosts and manages the entire software application, including the underlying infrastructure, middleware, and data. Customers access the application over the internet (typically via a web browser or API) without needing to install, maintain, or update anything locally. This model shifts all operational responsibility to the provider, making it the most 'hands-off' cloud service model for the customer.

Answer

A model where customers manage their own virtual machines and applications

Answer

A model where customers deploy their code on provider-managed platforms

Answer

A model where customers rent physical hardware from the provider

Question 12

Which statement best describes the concept of 'predictability' as a cloud benefit?

Accepted Answer

Confidence in consistent performance and the ability to forecast costs accurately. Predictability in cloud computing refers to the ability to rely on consistent performance and accurately forecast costs. This is achieved through Azure's Service Level Agreements (SLAs) that guarantee uptime and performance metrics, combined with tools like Azure Cost Management and Pricing Calculator that provide transparent, granular cost estimates. This allows organizations to plan budgets and resource allocation with confidence, avoiding unexpected expenses or performance degradation.

Answer

The ability to automatically replace failed resources without human intervention

Answer

The guarantee that all data is stored within a specific geographic region

Answer

The ability to deploy identical environments for development and production

Question 13

Which of the following is an example of a SaaS (Software as a Service) offering from Microsoft?

Accepted Answer

Microsoft 365. Microsoft 365 is a SaaS offering because it provides ready-to-use productivity applications (e.g., Word, Excel, Teams) accessed via a web browser or client, with Microsoft managing the underlying infrastructure, operating system, and application updates. The customer simply consumes the software without needing to install or maintain servers or runtime environments.

Answer

Azure Virtual Machines

Answer

Azure App Service

Answer

Azure SQL Database

Question 14

Which cloud benefit helps organizations meet workload demands during unexpected events like viral marketing campaigns without service degradation?

Accepted Answer

Elasticity. Elasticity is the correct answer because it refers to the ability of a cloud system to automatically provision and de-provision resources (such as compute instances, storage, or bandwidth) in real time to match fluctuating demand. During a viral marketing campaign, traffic spikes can be handled by scaling out resources dynamically, ensuring consistent performance without manual intervention or service degradation.

Answer

High availability

Answer

Security

Answer

Predictability

Question 15

When using Azure, a company wants to calculate the composite SLA for two services used together where both must be available. Service A has a 99.9% SLA and Service B has a 99.95% SLA. What is the composite SLA?

Accepted Answer

Approximately 99.85%. When two services must both be available (an 'and' scenario), the composite SLA is calculated by multiplying their individual SLAs. Service A at 99.9% (0.999) and Service B at 99.95% (0.9995) gives 0.999 × 0.9995 = 0.9985005, which rounds to approximately 99.85%. This reflects the increased probability of failure when multiple dependencies are required.

Answer

99.95% (the higher of the two SLAs)

Answer

99.9% (the lower of the two SLAs)

Answer

100% because Azure guarantees maximum availability