AZ-900 Practice Question 72: Fault Tolerance vs Disaster Recovery
Microsoft Azure Fundamentals (AZ-900) · Test 8 · question 2 of 10
A payment API runs on several instances behind a load balancer. When one instance crashes, the load balancer stops sending it traffic within seconds, and the remaining instances keep serving every request without interruption or a lost transaction. Which concept does this describe?
Show answer and explanation
Answer:
Fault tolerance is the ability of a system to continue operating, in some defined capacity, in the event of a failure, achieved through techniques such as redundancy, failover, and graceful degradation. Losing one instance out of several and continuing to serve every request without interruption is that definition in action: the remaining, redundant instances absorb the failure so the service as a whole never goes down.
Option A, disaster recovery, addresses a different class of problem: rare, high-impact events such as a regional outage, planned for with recovery point and recovery time objectives rather than day-to-day instance crashes. Option C, cost predictability, is about forecasting spend and has nothing to do with surviving a failure. Option D, elasticity, is about capacity tracking demand up and down, which is unrelated to a single instance crashing while demand stays flat.
Reference: What are Business Continuity, High Availability, and Disaster Recovery?