
Cloud service models describe how responsibility is divided between a cloud provider and its customer. The four models are Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS) and Function as a Service (FaaS), and they differ in how much the provider manages and how much is left to the customer. On the Cloud Digital Leader exam, a scenario usually describes what a company wants to control, or wants to avoid, and the right model follows from those requirements.
IaaS provides virtualized computing resources such as servers, storage and networking that an organization rents from the cloud provider instead of buying. The customer avoids the cost and complexity of owning and maintaining physical hardware, but it still controls and manages the operating systems, configurations and applications that run on those resources. Installing and managing applications remains the customer's job.
That combination is what makes IaaS distinct. The organization gets scalable resources without the burden of data center equipment, while staying in control of how those resources are managed at the software level. Of the four models, it is the one that leaves the operating system in the customer's hands.
PaaS abstracts more of the system. It provides a managed development and deployment environment where developers run their code without managing the underlying infrastructure, and it hides most of the operating system details. The provider handles the platform, and the customer focuses on the application.
Developers still have to write and deploy code on PaaS, and the operating system is largely hidden from them. That makes PaaS a middle ground, with less to manage than IaaS, but it is neither code-free nor a way to control the operating system directly.
SaaS delivers complete, fully built applications over the internet, such as email, document editing or messaging tools. The provider manages the infrastructure, updates and availability, and users simply log in, often through a web browser. The customer writes no code and installs or maintains nothing.
Of the four models, SaaS leaves the least responsibility with the customer. There is no development, hosting or system management on the customer's side, only the use of the application itself.
FaaS runs small, discrete pieces of code in response to events, and the customer manages no servers. It removes server management, but it still requires programming. That places it apart from both IaaS and SaaS. It is not a way to lease virtual servers, and it is not a finished application that users simply log in to.
The first property is who controls the operating system. IaaS leaves it with the customer, while the other three take it out of the customer's hands, with PaaS hiding most operating system details and FaaS and SaaS leaving no servers for the customer to manage. The second is whether the customer writes code. PaaS and FaaS both remove server management but still require programming, and SaaS is the only model in which the customer writes no code and manages no systems.
Between PaaS and FaaS, the difference is the unit of work. PaaS is an environment for developing and deploying applications, while FaaS runs small, discrete pieces of code when events occur. Both sit between the control that IaaS leaves to the customer and the finished product that SaaS delivers, and keeping that ordering in mind makes it easier to match a requirement to a model on the Cloud Digital Leader exam.
Our Cloud Digital Leader practice exams cover IaaS, PaaS, SaaS and FaaS alongside hybrid and multi cloud deployment models, with detailed explanations that drill these distinctions.