Choosing the NDFC physical three-node pND cluster deployment

300-610 By islever Editorial Last updated

Evaluate options for implementing automatic network deployment using Nexus Dashboard Evaluate options for interconnecting data centers with VXLAN EVPN Evaluate network management and redundancy strategies

Short answer

C Deploy a three-node physical Nexus Dashboard Fabric Controller (pND) cluster to provide supported active/active HA for the dual-site VXLAN fabric.

Question

An architect is designing a VXLAN solution for a client with limited experience in complex deployments and has recommended using the Nexus Dashboard Fabric Controller (NDFC) for automating the configuration and management of the solution. The NDFC is intended to manage a dual-site production data center environment in a physical appliance active/active high availability mode. Which NDFC deployment model must the architect select for the client?
  1. five-node vND cluster
  2. four-node pND cluster
  3. three-node pND cluster Correct Answer
  4. four-node vND cluster

Key Takeaways

The trap is assuming that a larger node count equals better HA; NDFC/Nexus Dashboard clusters exist only in odd, quorum-based sizes, so a four-node pND cluster is not a valid model.

This 300-610 question tests NDFC deployment models on Nexus Dashboard, where a dual-site production design on physical appliances with active/active HA requires the three-node pND cluster (answer C).

Common mistake: Option B (four-node pND cluster) is wrong because it honors the physical-appliance requirement but is not a supported cluster size — HA quorum needs an odd three- or five-node cluster, so four nodes cannot tolerate a node failure.

Expert Analysis

Why the physical-appliance wording matters The question fixes two constraints: the client is running a physical Nexus Dashboard appliance cluster, and it needs production HA across a dual-site design. That points to a pND model, which removes option A and option D (both vND, i.e. VM-based deployments) immediately. The remaining choice is how many physical nodes the cluster contains.

The attractive wrong answer: option B (four-node pND cluster) A four-node pND cluster sounds like the safest engineering instinct: it keeps the physical appliance requirement, it "adds redundancy," and it sits between the three- and five-node sizes candidates assume are available. It also matches the mental model of "more nodes = better HA for a dual-site production environment." This is where it breaks down. Nexus Dashboard, and therefore NDFC running on it, is deployed in fixed, quorum-based cluster sizes — three or five nodes — and HA/quorum relies on an odd-numbered cluster so a majority survives a node loss. A four-node cluster is not a supported NDFC deployment model; there is no valid majority once a node fails, so it cannot deliver the active/active HA the question demands. A comment in the study discussion makes exactly this point: a three-node cluster "can tolerate a Node failure."

Confirming option C Option C, the three-node pND cluster, satisfies every stated constraint: physical appliance, production HA with fault tolerance to a single node failure, and a supported NDFC deployment size for a dual-site fabric controller. Option A (five-node vND) is a legitimate cluster size but the wrong appliance type. Correct answer: C.

Exam Strategy

When a question names the appliance type (physical vs virtual) plus an HA mode, filter the options by appliance type first, then by supported cluster sizes: Nexus Dashboard/NDFC clusters are quorum-based and come in three or five nodes, never four. Treat any even node count as an automatic elimination unless the vendor explicitly documents it, then pick the smallest cluster that still tolerates the node failure implied by the HA requirement.

Frequently Asked Questions

Why can't NDFC run on a four-node Nexus Dashboard cluster?

Nexus Dashboard clusters are deployed in fixed, odd-numbered sizes (three or five nodes) so that quorum can be maintained after a node loss. A four-node cluster has no majority once a node fails, so it cannot provide the active/active HA NDFC requires.

When would a five-node NDFC cluster be the right choice?

A five-node cluster is selected for larger fabrics or higher node-failure tolerance, allowing the cluster to survive two simultaneous node failures. Note that the five-node option in this question is vND (virtual), which conflicts with the physical-appliance requirement.

What does pND vs vND mean for NDFC deployments?

pND runs NDFC on physical Nexus Dashboard appliances, while vND runs it on virtual machines in a hypervisor. Both are valid NDFC platforms, but the question explicitly specifies physical appliance HA, which restricts the choice to pND models.

More 300-610 FAQ →

Community Votes

C
100%

100% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 1 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 1

  1. garza_javi 👍 1 Answered C
    Three node cluster, can tolerate a Node failure

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.

Cisco UCS Fabric Interconnect Storage Port Configuration

S-grade

Cisco UCS Director for Data Center Orchestration

S-grade

UCS Fabric Interconnect FC Storage Port Configuration

S-grade

FCoE Native VLAN Restriction and VSAN Configuration

S-grade

← Back to 300-610 Study Guide