Comprehensive and Detailed Explanation (HCIA–Cloud Computing aligned):
In FusionCompute architecture, CNA and VRM have clearly defined deployment models.
Option A is false. CNA is deployed only on physical servers (compute nodes). It is responsible for interacting directly with hardware resources and cannot run inside a VM.
Option B is true. When VRM is deployed on physical servers, active and standby VRM nodes must be installed on separate physical servers to ensure high availability.
Option C is true. When VRM is deployed on virtual machines, the active and standby VRM VMs must be placed on different hosts within the management cluster to avoid single points of failure.
Option D is true. VRM supports deployment on either physical servers or virtual machines, providing flexible management options.
Therefore, the correct answers are B, C, and D.
