Q: 10
A large corporation recently experienced a power outage at one of its primary data centers resulting
in service disruption for customers in that region. An administrator is tasked to assess the current
infrastructure and propose a plan to improve resiliency.
Current configuration:
Single-site vSAN Express Storage Architecture (ESA) cluster
12 hosts
Cluster resource utilization (CPU, memory, and storage) is under 30%
Which solution would improve resiliency and minimize service disruption in data center outages with
a recovery point objective (RPO) of zero without requiring additional hosts?
Options
Discussion
A , fault domains (D) are good for local failures but don't cut it for a real site outage. Stretched cluster with option A is how you get RPO zero across sites, and you don't need extra hosts. If I'm missing something, let me know.
D isn't going to help with full site loss. A is right.
Why wouldn't D work here if fault domains are in the same site anyway?
Answer: Saw a similar scenario in a mock test, and A is correct. vSAN Stretched Cluster lets you use the existing hosts, splits them across two locations, and enables synchronous replication for RPO zero. The rest don't meet the RPO or hardware constraints. Feel free to disagree if you found something different!
B tbh
Had something like this in a mock, pretty sure A fits best. Splitting the 12 hosts into two sites with vSAN stretched cluster is the only way here to get synchronous replication and RPO zero without more hardware. The others don't guarantee zero data loss. If anyone's seen it answered otherwise, let me know.
I don't think B fits here. A is the only one with synchronous replication between sites, which is what you need for true RPO zero. Plus, you can split the existing 12 hosts so no need for new hardware. Pretty sure that's what VMware wants in this case.
Option A. since stretching the cluster lets you keep RPO at zero without extra hardware. Stretched vSAN works for this scenario.
Does the question mean "best" resiliency, or just any improvement? If the RPO wasn’t strictly zero, then maybe B could work, but with RPO=0 I think A is the only option.
Its A here, since only a stretched vSAN cluster with hosts in both sites gets you synchronous replication and true RPO zero. D (fault domains) is for rack or local issues, not full site outage. Pretty sure that's what VMware wants.
Be respectful. No spam.
