What selection is the key design benefit provided by a dedicated Edge Cluster VM or Bare Metal? (Choose the best answer.)
What selection is the key design benefit provided by a dedicated Edge Cluster VM or Bare Metal? (Choose the best answer.)
A dedicated Edge Cluster VM or Bare Metal solution ensures predictable network performance because it eliminates contention for resources typically seen in a shared virtual environment. With dedicated hardware, the network services such as routing and firewalling can operate without interference from other workloads, leading to consistent performance and reliability. This predictability is crucial for applications that demand stable and reliable network performance.
Answer B is correct. https://docs.vmware.com/en/VMware-Validated-Design/services/deployment-of-nsx-t-edge-nodes-on-bare-metal-hardware-for-vmware-cloud-foundation-40/GUID-563E93F0-65C8-4649-B62F-9AFE89B08B50.html https://vxplanet.com/2019/06/13/nsx-t-edges-baremetal-vs-vm-comparison/
B is correct
The answer is undoubtedly A
I see this question as more tricky than it would seem on the first sight - you see word performance and go with option B immediately, but the wording is different: PREDICTABLE network performance So what does make bare metal edge MORE predicatble than VM performance wise? In my opinion the fact that you have more raw compute power and thus more performance dosn't make it predictable. Reading VVD https://docs.vmware.com/en/VMware-Validated-Design/services/deployment-of-nsx-t-edge-nodes-on-bare-metal-hardware-for-vmware-cloud-foundation-40/GUID-563E93F0-65C8-4649-B62F-9AFE89B08B50.html there are more arrows for the difference about manageability rather than performance. So I'm more inclined for answer A rather than B (C and D are plain wrong).
it doesnt need to contend with other vm for resources.
That is true and a good point, but it is quite speculative - you can for sure design the solution so there won't be any CPU contention. Looking on it from your perspective maybe more predicatble performance is in the non-existent overhead of the hypervisor layer in the case of bare-metal. Still the VVD is quite clear (at least I see it there).
there are differences in manageability between the Bare Metal and virtual edge nodes true, but nothing is a reduced overhead with Bare Metal Edge nodes, it even adds complexity to the design and deployment, B is the correct answer.
B is correct
B. Predictable network performance A dedicated Edge Cluster ensures that the network performance for edge services, such as routing and firewalling, can be consistently maintained and doesn't compete with compute workloads, reducing the risk of performance variations and ensuring predictable network performance.
Answer B. Faster Convergence with BGP peering. 3 Seconds for Edges, 500ms for Bare Metal