Cloud Service Models Questions and Answers

Answer: A

Correct answer: virtual machines, storage and networking

Infrastructure as a service delivers the raw building blocks - compute, storage and network - and everything above the hypervisor is the customer’s: the operating system, its patching, the runtime, the middleware and the application. A development platform with a runtime is platform as a service, and a finished application is software as a service.

Answer: B

The stack is handed over progressively, and the customer's remaining responsibility shrinks with each step of the ladder. Control shrinks alongside it, which is why the highest available model is not automatically the right one to pick.

Answer: C

Correct answer: Both A and R are true, and R is the correct explanation of A

Both are true and the reason describes the mechanism. Each step up the stack - infrastructure, platform, software - moves another layer from the customer’s side of the line to the provider’s. Less to operate necessarily means less to decide, so the trade is intrinsic rather than incidental, and the right level is whichever gives away the control that is not worth keeping.

Answer: D

Code is deployed to a managed runtime and the platform handles patching and scaling. The customer still writes and owns the application and its data. Patching moves to the provider. Scaling is handled by the platform as well.

Answer: A

The first two are correct. The third is exactly what the model removes, and it is one of the main reasons for choosing it over raw infrastructure. Code and data stay with the customer. Runtime versions are chosen but not maintained.

Answer: B

Correct answer: a hosted email or relationship product

Software as a service is a finished application the customer simply uses, with no infrastructure, runtime or application maintenance visible - hosted email and CRM being the standard examples. A block volume and a rented virtual machine are infrastructure, and a managed Kubernetes cluster is a platform on which the customer still deploys and runs their own software.

Answer: C

Correct answer: patching the application server

Patching is the provider’s in software as a service; it is precisely what the customer is paying not to do. What remains theirs is everything specific to their own use: how the application is configured, the data they put into it, and who is allowed access. The last of these is the commonest cause of a software-as-a-service breach, since a misconfigured share or an over-privileged account is a customer fault, not a provider one.

Answer: D

Both are true and the reason gives the necessity, and misconfigured access is among the commonest causes of data exposure in hosted applications. Only the customer knows its own staff. Permissions cannot be delegated upward.

Answer: A

The unit of deployment is a function and the platform provides the runtime on demand. Billing is by invocation and duration rather than by reserved capacity. The function is the unit deployed. Idle functions cost nothing at all.

Answer: B

The first two are correct. The third is a common misreading of the word serverless, since servers are certainly involved but they are none of the customer's concern. The name describes the model. Servers run beneath as they always did.