The Administering Windows Server Hybrid Core Infrastructure guidance explains two Windows container isolation modes: process isolation (Windows Server containers) and Hyper-V isolation. With process isolation, containers share the host’s kernel , so they run as ordinary processes on the host. With Hyper-V isolation, each container runs inside a lightweight VM with its o wn kernel ; this provides a hard isolation boundary and prevents kernel sharing with other containers. The curriculum further notes that Linux containers on Windows require Hyper-V isolation because they cannot share the Windows kernel.

Applying these rules:

Container1 “must NOT share a kernel with other containers.” The only mode that prevents kernel sharing is Hyper-V isolation, so Container1 must use Hyper-V isolation.

Container2 is a Linux container. On a Windows Server container host, Linux containers are run via a utility VM and therefore require Hyper-V isolation.

Container3 is a Windows container that needs a static IP address. The networking modules describe that Windows containers can obtain static IPs when you use supported networks (for example, l2bridge/transparent) regardless of isolation mode; therefore a Windows workload like a database can run in either process or Hyper-V isolation, depending on the isolation/security you want.

Thus, the correct selections are: Hyper-V isolation only for Container1 and Container2, and Hyper-V isolation or process isolation for Container3.