You need to identify which cloud service models place the most responsibility on the customer...

Microsoft’s shared responsibility guidance explains that the customer’s responsibility decreases as you move from on-premises to SaaS. In an on-premises datacenter, “you own the whole stack—applications, data, runtime, middleware, OS, virtualization, servers, storage, and networking.” With IaaS, the cloud provider operates the physical datacenter and virtualization, while “you’re responsible for configuring and managing the guest OS, network controls, identity, applications, and data.” With PaaS, the provider operates more of the stack so that “the cloud provider manages the platform (OS, middleware, and runtime) and you focus on your applications and data.” Finally, with SaaS, responsibility is minimized for customers because “the service provider manages the application and underlying infrastructure; customers primarily manage identity, data, and access/usage.”

These Microsoft Learn statements map directly to the requested order—from most customer responsibility (on-premises) to least (SaaS)—with IaaS and PaaS in between, reflecting the progressive shift of operational and security controls from the customer to the cloud provider as the service model moves up the stack.