A production application runs on three Azure VMs in the same region. The business requires the service to stay available if one entire datacenter in the region becomes unavailable because of a power or network outage. Which configuration best meets the requirement?

Place the VMs in the same availability set.

An availability set protects against host and maintenance failures, not a full datacenter outage.

Use a proximity placement group for the VMs.

A proximity placement group optimizes latency, but it does not provide datacenter-level fault isolation.

Attach the VMs to the same Azure Load Balancer backend pool.

A load balancer distributes traffic, but it does not provide physical redundancy by itself.

What does this AZ-104 question test?

OSPF neighbours must agree on key parameters.

What is the correct answer to this question?

The correct answer is: Deploy the VMs across availability zones. — Availability zones are designed for this type of requirement because they span physically separate datacenters inside a region. If one zone suffers a power, cooling, or network failure, workloads in the other zones can continue serving traffic. An availability set is useful for host or rack maintenance resilience, but it does not protect against an entire datacenter failure. The load balancer or placement group can support the design, but they do not provide the fault isolation asked for here. Why others are wrong: An availability set only spreads VMs across update and fault domains within the same datacenter area. A proximity placement group improves co-location and latency, which is the opposite of the needed separation. A load balancer helps traffic distribution but cannot keep the application alive when the underlying compute location is lost.

What should I do if I get this AZ-104 question wrong?

Then try more questions from the same exam bank and focus on understanding why the wrong options are tempting.