In a 20-node ESXi deployment, the customer needs additional resiliency to survive the loss of two nodes concurrently. Which configuration should the consultant use to ensure that the appropriate resources are available?

To survive the concurrent loss of two nodes, the cluster's storage must be configured with Replication Factor 3 (RF3), which maintains three copies of all data. This ensures data availability if two nodes fail. For compute resource availability in an ESXi environment, VMware High Availability (HA) admission control must be configured to reserve sufficient resources for restarting the failed nodes' VMs. In a 20-node cluster, the loss of two nodes represents 10% of the total cluster resources (2 / 20 = 0.10). Therefore, HA admission control should be set to reserve 10% of the cluster's CPU and memory resources to guarantee successful VM failovers.

B. This option uses RF2, which can only tolerate a single node failure, and reserves insufficient HA resources (5%) for a two-node failure scenario.
C. While this option correctly specifies RF3, the HA admission control of 5% is insufficient, as it would only reserve enough resources to handle the failure of one node (1/20).
D. This option uses RF2, which cannot tolerate the failure of two nodes concurrently, making the HA setting irrelevant for the customer's requirement.

1. Nutanix. (2023). vSphere Administration Guide for AOS 6.7. Section: Configuring High Availability and DRS. "To tolerate n host failures, set the Percentage of Cluster Resources Reserved policy to (n / totalhosts) 100. For 2 failures in a 20-node cluster, this is 10%."

2. Nutanix. (2023). Prism Web Console Guide v6.7. Section: Updating Cluster Details. "To ensure availability after two simultaneous node failures, the cluster must be configured with redundancy factor 3."

3. Nutanix Bible. (2023). Section: Data Protection - Replication Factor. "Replication Factor 3 (RF3) provides the ability to tolerate up to two simultaneous component (e.g., node or disk) failures."