Mixed practice exam 1

Explanation

Answer: C. "Go global in minutes" is AWS's wording for the advantage that the global infrastructure lets you deploy an application in multiple Regions around the world with just a few clicks, which provides lower latency and a better experience for customers at minimal cost. Achieving the same reach on premises would mean leasing space, shipping hardware, and hiring staff in each country.

• A. Trade fixed expense for variable expense describes paying only when you consume resources instead of investing heavily up front in data centers and servers. It is a financial advantage, not a geographic one.
• B. Stop guessing capacity is about eliminating the need to forecast infrastructure capacity before deployment. It addresses how much you provision, not where you provision it.
• D. Benefit from massive economies of scale explains why AWS pay-as-you-go prices are lower than a single customer could achieve alone. It describes price levels, not global reach.

Explanation

Answer: C. Elasticity is the ability to acquire resources as you need them and release them when you no longer need them, automatically and in near real time. It is what makes a consumption-based cost model work in practice: capacity tracks demand instead of being fixed at the peak.

• A. Loose coupling is an architectural practice of reducing dependencies between components so that one failing does not cascade. It is a design technique rather than a capacity behavior.
• B. Fault tolerance is the ability of a system to keep operating correctly even when one or more of its components fail. Redundancy delivers that, not automatic scaling in and out.
• D. Durability describes the likelihood that stored data survives without loss or corruption over time. It is a property of storage, not of capacity that grows and shrinks.

Explanation

Answer: C. AWS describes this as "increase speed and agility": because new resources are only a click away, the time to make them available to developers drops from weeks to minutes. Since the cost and time to experiment fall dramatically, teams can try more ideas and discard failures cheaply, which is the essence of agility.

• A. Data sovereignty is about which jurisdiction data resides in, which is governed by the Region you choose rather than by how quickly resources are created.
• B. Blast radius refers to how much of a system is affected when a failure or a mistake occurs. It is a resilience and isolation concept, not a provisioning speed concept.
• D. Durability concerns how reliably data is preserved over time. Faster provisioning has no bearing on whether stored objects survive.

Explanation

Answer: B. High availability means the workload keeps serving requests, with as little downtime as possible, even when individual components fail. It is achieved by removing single points of failure, typically by running redundant components across multiple Availability Zones and routing traffic only to healthy ones.

• A. Maximum sustained throughput is a measure of performance and capacity, not of whether the system stays reachable during a failure.
• C. Encrypting data in transit and at rest is a security control that protects confidentiality. It does not keep a workload running.
• D. That describes recoverability or a recovery point objective, which is a backup and disaster recovery concern rather than continuous uptime.

Explanation

Answer: B. AWS owns and operates a worldwide footprint of Regions, Availability Zones, and edge locations. A customer can select and deploy into that infrastructure on demand, so entering a new market becomes a configuration decision instead of a construction project. This is the practical meaning of global reach and speed of deployment.

• A. Most AWS services are Regional, and the customer chooses the Region. Location remains a deliberate architectural decision rather than something abstracted away.
• C. AWS does not replicate customer data between Regions automatically. Regions are isolated, and any cross-Region copying is something the customer explicitly configures.
• D. Latency depends on physical distance and network path, so it varies by Region and by user. Reducing latency is precisely why customers deploy closer to their users.

Explanation

Answer: D. AWS defines cloud computing as the on-demand delivery of IT resources over the internet with pay-as-you-go pricing. The two ideas that matter are on demand, meaning you provision what you need when you need it without a procurement cycle, and pay-as-you-go, meaning you are charged for consumption rather than for owned capacity.

• A. Managed hosting of a customer's own physical servers still leaves the customer owning fixed capacity that must be forecast and purchased in advance.
• B. A perpetual license installed locally is the traditional model that cloud consumption replaces or supplements. It is a software purchasing arrangement, not a computing delivery model.
• C. Running your own hardware in a colocation facility is still a capital purchase of fixed capacity. You gain a better building but none of the on-demand or consumption-based characteristics.

Explanation

Answer: B. AWS's wording is that you eliminate guessing on infrastructure capacity needs: instead of committing to a size before deployment and then either sitting on expensive idle resources or hitting a ceiling, you access as much or as little capacity as you need and scale up and down with only a few minutes' notice.

• A. AWS accounts have service quotas, and many can be raised on request, but they are not unlimited. The advantage is about avoiding advance commitment, not about the absence of limits.
• C. AWS does not forecast a customer's demand or provision on their behalf. The customer configures scaling policies; AWS supplies the capacity when those policies call for it.
• D. There is no refund for unused provisioned capacity. Avoiding that waste is exactly why elasticity matters.

Explanation

Answer: B. Fault tolerance is built-in redundancy that allows a workload to keep functioning correctly through the failure of one or more components, ideally without the user noticing. It is a stronger property than high availability, which accepts brief disruption while the system recovers or fails over.

• A. Restricting source IP addresses is a network access control. It reduces exposure but does nothing to keep the workload running when a component fails.
• C. Encrypting internal traffic protects confidentiality and integrity. An encrypted system with a single point of failure is still not fault tolerant.
• D. Deleting data on an age-based schedule is a lifecycle and retention practice used to manage storage cost, not a resilience property.

Explanation

Answer: B. The waste comes from owning peak-sized capacity all year. Elasticity removes it: the retailer scales out for the four peak days and scales back in afterwards, paying for the extra capacity only while it is running. This is the operational mechanism behind both "stop guessing capacity" and the consumption-based cost model.

• A. Fault tolerance keeps the workload running through component failures. It is valuable during the peak but it is not what eliminates the year-round cost of idle peak capacity.
• C. Data sovereignty concerns the jurisdiction in which data is held. It is a compliance benefit and does not change capacity economics.
• D. Durability protects stored data from loss. It has no effect on how much compute capacity the retailer pays for across the year.

Explanation

Answer: A. Agility is the benefit AWS attaches to resources being only a click away: because the cost and time to experiment fall dramatically, failed experiments become cheap and teams can run more of them. Provisioning five environments and terminating four costs only the hours they ran.

• B. Economies of scale explain why AWS unit prices are lower than a customer could achieve alone. That lowers the bill but is not the mechanism that makes rapid experimentation possible.
• C. Durability concerns whether stored data survives. It says nothing about how quickly a team can stand up and tear down test environments.
• D. High availability matters for production workloads that must stay up. Short-lived prototypes typically do not need multi-Availability Zone redundancy at all.