The VCF 5.2 design must ensure the business application (two VMs) remains available during

business hours (9 AM - 5 PM weekdays) and is protected from storage I/O disruptions in a

consolidated architecture with a single six-host cluster using vSAN. The goal is to mitigate risks to the

application’s performance and availability. Let’s evaluate each option:

Option A: Use resource pools to apply CPU and memory reservations on the business application

virtual machines

Resource pools with reservations ensure CPU and memory availability, which could help

performance. However, the application’s sensitivity is to storage I/O, not CPU/memory, and the

availability requirement (business hours) isn’t directly addressed by reservations. While useful, this

doesn’t fully mitigate the primary risks identified, making it less optimal.

Option B: Implement FTT=6 for the business application virtual machines

This is incorrect and infeasible. In vSAN, Failures to Tolerate (FTT) defines the number of host or disk

failures a storage object can withstand, with a maximum FTT dependent on cluster size. FTT=6

requires at least 13 hosts (2n+1 where n=6), but the cluster has only six hosts, supporting a maximum

FTT=2 (RAID-5/6). Even if feasible, FTT addresses data redundancy, not runtime availability or I/O

sensitivity during business hours, making this irrelevant to the stated risks.

Option C: Perform ESXi host maintenance activities outside of the stated business hours

This is the correct answer. In a vSAN-based VCF cluster, ESXi host maintenance (e.g., patching,

reboots) triggers data resyncs and VM migrations (via vMotion), which can impact storage I/O

performance and potentially cause brief disruptions. The application’s sensitivity to storage I/O and

its availability requirement (9 AM - 5 PM weekdays) mean maintenance during business hours poses

a risk. Scheduling maintenance outside these hours (e.g., nights or weekends) mitigates this by

ensuring uninterrupted I/O performance and availability during critical times, directly addressing the

customer’s needs.

Option D: Replace the vSAN shared storage exclusively with an All-Flash Fibre Channel shared storage

solution

This is incorrect. While an All-Flash Fibre Channel array might offer better I/O performance, VCF’s

consolidated architecture relies on vSAN as the primary storage for management and workload

domains. Replacing vSAN entirely contradicts the chosen architecture and introduces unnecessary

complexity and cost. The sensitivity to storage I/O changes doesn’t justify abandoning vSAN,

especially since All-Flash vSAN could meet performance needs if properly tuned.

Option E: Use Anti-Affinity Distributed Resource Scheduler (DRS) rules on the business application

virtual machines

Anti-Affinity DRS rules ensure the two VMs run on separate hosts, improving availability by avoiding

a single host failure impacting both. While this mitigates some risk, it doesn’t address storage I/O

sensitivity (a vSAN-wide concern) or guarantee availability during business hours if maintenance

occurs. It’s a partial solution but less effective than scheduling maintenance outside business hours.

Conclusion:

The best design decision is to perform ESXi host maintenance activities outside of the stated business

hours (Option C). This directly mitigates the risk of storage I/O disruptions and ensures availability

during 9 AM - 5 PM weekdays, aligning with the customer’s requirements in the VCF 5.2 consolidated

architecture.

Reference:

VMware Cloud Foundation 5.2 Architecture and Deployment Guide (Section: Consolidated

Architecture Design)

VMware vSAN 7.0U3 Planning and Deployment Guide (integrated in VCF 5.2): Maintenance Mode

Considerations

VMware Cloud Foundation 5.2 Planning and Preparation Guide (Section: Availability and

Performance Design)