DOP-C02 Resilient Cloud Solutions Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✓

  Deploy resources across multiple Availability Zones

  Why this is correct

  Deploying across multiple Availability Zones (AZs) provides infrastructure-level fault isolation because each AZ has independent power, cooling, and network access. This architecture ensures that an outage in one AZ does not take down the entire workload, supporting a higher availability SLA. For example, running EC2 instances in two or more AZs with an Application Load Balancer allows traffic to continue to healthy AZs even if one becomes isolated.

• ✗

  Use a single large instance instead of multiple smaller ones

  Why it's wrong here

  Using a single large instance consolidates compute into one server, creating a single point of failure: if the instance fails due to hardware degradation, operating system crash, or underlying host maintenance, the workload becomes completely unavailable. Additionally, vertical scaling has hard limits for instance sizes and purchase types, and the instance cannot provide redundancy for itself. A horizontally scaled fleet of smaller instances behind a load balancer avoids this risk and enables rolling deployments.

• ✓

  Use health checks to automatically replace unhealthy resources

  Why this is correct

  Health checks continuously monitor the status of resources (such as EC2 instances, containers, or endpoints) using protocols like HTTP/TCP, and trigger automated actions when they mark a resource as unhealthy. With Auto Scaling groups, an unhealthy instance is terminated and replaced automatically; with Route 53, traffic is redirected away from failing endpoints. This automated remediation reduces mean time to recovery (MTTR) and enables unattended operational recovery, which is essential for resilience.

• ✗

  Disable automated backups to reduce latency

  Why it's wrong here

  Automated backups are primarily for data durability and disaster recovery, not for affecting runtime latency; AWS snapshot-based backups (e.g., EBS snapshots, RDS automatic backups) are asynchronous and run in the background with negligible interference. Disabling backups to gain a tiny performance margin would expose your system to permanent data loss if the underlying storage fails or a logical error occurs, and you would have no point-in-time recovery capability.

• ✗

  Deploy resources in a single Availability Zone

  Why it's wrong here

  Placing all resources in a single Availability Zone means the entire application shares one physical failure domain; an AZ failure due to power loss, cooling malfunction, or network fiber cuts can cause a complete outage. This violates the core resilience principle of avoiding single points of failure at the infrastructure level. Distributing workloads across AZs is a foundational best practice for meeting RTO/RPO targets, as no single in-region event can take down the system.

Go deeper

Related to this question