Controlled Layer 2 Between VMs on the Same Hypervisor

350-401 By islever Editorial Last updated

Describe device virtualization technologies

Short answer

AB Use a hypervisor-provided virtual switch or a VM-based virtual switch to provide controlled Layer 2 connectivity between VMs on the same hypervisor.

Question

Which two actions provide controlled Layer 2 network connectivity between virtual machines running on the same hypervisor? (Choose two.)
  1. Use a virtual switch provided by the hypervisor. Correct Answer
  2. Use a virtual switch running as a separate virtual machine. Correct Answer
  3. Use VXLAN fabric after installing VXLAN tunneling drivers on the virtual machines.
  4. Use a single routed link to an external router on stick.
  5. Use a single trunk link to an external Layer2 switch.

Key Takeaways

The question tests whether you identify virtual switches—hypervisor-integrated or VM-based—as the Layer 2 control point for same-host VMs, while the external trunk is the main trap.

This Cisco ENCOR 350-401 item tests virtual switching for VMs on one hypervisor. The controlled Layer 2 path is provided by a hypervisor virtual switch or a VM-based virtual switch (A, B), not by an external trunk or routed link.

Common mistake: Selecting E because a trunk to an external Layer 2 switch sounds like a controlled VLAN design; however, it does not create the intra-hypervisor switching path required between VMs on the same host and would unnecessarily hairpin local traffic.

Expert Analysis

The same-hypervisor constraint is decisive. The question is not asking how to extend a VM VLAN into the physical data center; it asks how two VMs on one host obtain controlled Layer 2 connectivity. In that scope, the Layer 2 control point is a virtual switch inside the hypervisor, not an external router or external switch.

Why E is the attractive trap. Option E sounds right because a single trunk to an external Layer 2 switch is a familiar way to carry VLANs and apply switch-port policy. It satisfies the words "controlled" and "Layer 2," and discussion comments imagine two different vSwitches that need outside help. The single point where it fails is the question's locality: a trunk uplink does not create the intra-host virtual switching path between the VMs, and using it for same-host traffic would hairpin frames out to the physical switch and back. The question also does not state that the VMs are on separate vSwitches, so that external-switch scenario is an assumption.

The correct pair. Option A supplies the normal hypervisor-integrated vSwitch, where port groups, VLANs, and virtual-switch policy provide controlled Layer 2 connectivity. Option B supplies the other virtual-switch model: a virtual switch deployed as a separate VM, such as a Nexus 1000V-style virtual switch control component, which can also provide policy-controlled Layer 2 for VMs. Option D is Layer 3, and option C is an overlay mechanism that is unnecessary for same-host switching. Therefore the correct answers are A and B.

Official Reference

Exam Strategy

When a virtualization question specifies "same hypervisor," eliminate options that send local VM-to-VM traffic through external routers or switches; those solve inter-host or external connectivity. Look for the virtual switch as the Layer 2 control point, and distinguish a hypervisor-integrated vSwitch from a virtual switch appliance running as a separate VM. Also watch the Layer 2 versus Layer 3 keyword: a router-on-a-stick is always a Layer 3 distractor.

Frequently Asked Questions

Can VMs on the same hypervisor communicate through a hypervisor virtual switch without an uplink?

Yes. If both VMs attach to the same virtual switch or port group and external connectivity is not required, the vSwitch switches frames locally between their virtual NICs.

Why is a trunk to an external Layer 2 switch not the primary answer?

It can extend VLANs and provide external control, but it does not replace the intra-host virtual switching needed for same-hypervisor VM-to-VM traffic; it would hairpin local traffic unnecessarily.

What does a virtual switch running as a separate VM mean?

It refers to a virtual switch deployed as a VM-based component, such as a Nexus 1000V-style virtual switch control VM, which can provide policy-controlled Layer 2 switching for virtual machines.

More 350-401 FAQ →

Community Votes

A
51%
E
25%
B
24%

51% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 63 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 30 of 63

  1. Pb1805 👍 29
    Shouldn't it be A & E?
  2. xziomal9 👍 26
    The correct answer is: A. Use a virtual switch provided by the hypervisor. B. Use a virtual switch running as a separate virtual machine.
  3. ahmedox 👍 6
    i think the answers are A & B, the key here is: "Layer 2 connectivity" and "on the same Hypervisor" so we can provide Layer 2 connectivity inside the Hypervisor using Vswitch without going out of the Physical server.
  4. AlexLAN 👍 5
    Point is - "Layer 2 Connectivity", so D & E are both wrong - we can't create an L2 connection between two hosts on the same HV via Routed or Trunk link - it will be L3 connection. So, the right answer could be: A (vSwitch) & B (via Nexus 1000V).
  5. kewokil120 👍 5
    Sure, E can work, but what does it do that a virtual switch cant? VMs can talk with no external switch involved. with B, its possible with Nexus1000v, or CSR1000v A.B for me.
Show 25 more comments
  1. Miguex125 👍 5
    For this parte "between virtual machines running on the same hypervisor", doesnt make sense use a trunk link to an external switch. Could be A,B the answers
  2. zzmejce 👍 4
    This config doesn’t need external switch. Everything stays inside the physical server.
  3. MaxwellJK 👍 4
    A y B, el truco aqui esta que preguntan por dentro del mismo host
  4. Kanenas 👍 4
    A&E are correct
  5. Mimimimimi 👍 3
    I believe the answer should be A&B: A: Each VM could use the vSwitch, provided by its hypervisor. -> vSwitch uses its own local port configuration. B: DVS, Distributed Virtual Switches. A DVS which is associated with multiple ESX hosts, use the same port configuration across all hosts. -> Easier to manage when there are multiple ESX' in a datacenter. Source 1: https://www.cisco.com/c/dam/en/us/products/collateral/switches/nexus-1000v-switch-vmware-vsphere/at_a_glance_c45-532467.pdf Source 2: https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/vm_fex/vmware/gui/config_guide/2-1/b_GUI_VMware_VM-FEX_UCSM_Configuration_Guide_2_1/b_GUI_VMware_VM-FEX_UCSM_Configuration_Guide_2_1_chapter_0110.pdf
  6. HungarianDish_111 👍 3
    According to https://kb.vmware.com/s/article/2043160 "When using virtual switches without any uplink adapters, the virtual machines must reside on the same host in order to communicate with each other." -> without any uplink adapters to physical switches -> must reside on the same host (representation of physical server) which runs a bare-metal hypervisor Imho, opotion A and B can be applied in this scenario. -> virtual machines running on the same hypervisor
  7. RREVECO 👍 3
    Alternative "E" is required ref https://www.vmware.com/products/vsphere-hypervisor.html Pool computing and storage resources across multiple physical hosts. Have centralized management of multiple hosts through VMware vCenter Server. (for that connection an external switch is needed, connecting between different hosts with a vswitch is not justified)
  8. Asymptote 👍 3
    AE Point is that the question didn't mention how many virtual switches the virtual machines are connected to. For example: you have 2 virtual switches S1and S2, 2 VMs PC1 and PC2 PC1 is connected to S1 PC2 is connected to S2, if PC1 want to talk to PC2, there are no ways for both switches S1 and S2 talk directly, we have to use method E.
  9. hprc2002 👍 3
    The way the question is asked, the answers are A&B. The 2 VMs run on the same hypervisors. No external device needed the would require the use of L3
  10. kewokil120 👍 3
    AB is correct. Why punt it out of the domain unless it has too. Nexsu1000v is B
  11. brightsyds 👍 3
    Definitely A&B
  12. JaroslavS 👍 2
    There are 3 critical points in the question, it is: 1. controlled 2. Layer 2 3. same hypervisor This means, that D is not correct - it is layer 3. C is probably not correct, because using VXLAN does not help in communication inside hypervisor. I think A is not correct, because default vswitch usually does not provide the tools to controll (I mean filter) traffic between VMs. So B is absolutely correct, because there are several such switches running inside VM, that offer conntrolling the traffic. E is in question - I can imagine that in ACI, where I can switch and controll traffic between VLANs on the trunk, still being switched on L2. Not sure if other switch can do that.
  13. black2jm 👍 2
    I think it is A&E, but what about B? A. Use a virtual switch provided by the hypervisor. --> 2 VMs hooked to same vSwitch. Correct B. Use a virtual switch running as a separate virtual machine. --> Would this be considered something like Nexus 1k, or does 1k architecture fall under answer A. E. Use a single trunk link to an external Layer2 switch. --> Seems to fit. I could design it to work.
  14. Nickplayany 👍 2
    A and B are the correct answers. VXLAN fabric, a single routed link to an external router on stick, and a single trunk link to an external Layer 2 switch are all network connectivity solutions, BUT THEY ARE NOT specifically designed for providing controlled Layer 2 network connectivity between virtual machines running on the same hypervisor.
  15. JackyChon 👍 2
    The two actions that provide controlled Layer 2 network connectivity between virtual machines running on the same hypervisor are: A. Use a virtual switch provided by the hypervisor. B. Use a virtual switch running as a separate virtual machine.
  16. rocketplus 👍 2
    page 12 from OCG: "Thanks to the introduction of virtualization, some servers run a hypervisor for the operating system and contain a virtualized switch with different VLANs. These servers also provide connectivity via a trunk port".
  17. davdtech 👍 2
    In a real-world environment, VM is connected to a virtual switch then to a Layer 2 switch. D is incorrect, Connecting it to a router on a stick is l3 connectivity, controlled L2 connectivity is performed by a switch
  18. Burik 👍 2
    A and B don't need additional hardware and cabling, the hypervisor can take care of both scenarios. In a question related to virtualization within the same hypervisor why in the world would you go with E?
  19. pouriya 👍 2
    If 2 VMs needs to comunicate on L2 the virtual switch handels it but if they are on saprated Vlans tge trafic needs to be handeled on l3 sw or router... So the provided answers are correct
  20. KZM 👍 2
    For the Layer 2 connectivity between the VMs running on the same hypervisor host, no need to through the underlay Network (Physical Switch), I think.
  21. Eddgar0 👍 2
    Should be A and B, as is refering for Layer 2 on same host so, only A and B apply on this scenario.
  22. jacop 👍 2
    what if 2 diffrent vlans needs to communicate with each other on VMs ?
  23. MinC 👍 2
    I think the answer D: L2 networks connect together by L3 device.
  24. eff3 👍 2
    AB is better. E seems right but is missing too much details
  25. kthekillerc 👍 2
    Provided answers are correct.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Practice All 350-401 Questions

Access 218 questions with complete answers and detailed explanations.

View Full 350-401 Practice Test →

← Back to 350-401 Study Guide