AWS Certified SysOps Administrator – Associate — All Questions

A web tier must add capacity automatically when average CPU exceeds 60% and remove it when demand drops. Which configuration achieves this?

• a.A single large instance with detailed monitoring
• b.An Auto Scaling group with a target tracking policy on average CPU✓
• c.Manually launching instances during peaks
• d.A Reserved Instance purchase

A target tracking scaling policy adjusts the Auto Scaling group to keep average CPU near the target, adding and removing instances automatically. A single instance cannot scale horizontally, manual launching is not automatic, and Reserved Instances are a pricing commitment, not a scaling mechanism.

To protect a production database against an Availability Zone failure with automatic failover, which RDS option should be used?

• a.Multi-AZ deployment with a synchronous standby✓
• b.A single-AZ instance with frequent snapshots
• c.A read replica in the same AZ
• d.Larger instance class

RDS Multi-AZ maintains a synchronous standby in another Availability Zone and fails over automatically if the primary fails. Snapshots aid recovery but do not provide automatic failover, a same-AZ read replica does not survive that AZ's failure, and a larger instance does not add availability.

An organization wants one service to centrally schedule and enforce backups across EBS, RDS, and DynamoDB with retention policies. Which service fits?

• a.EBS snapshots created by a cron job
• b.S3 versioning
• c.CloudFormation
• d.AWS Backup✓

AWS Backup centralizes backup scheduling, retention, and vault policies across multiple services from one place. Cron-based snapshots cover only EBS and add operational burden, S3 versioning protects objects not databases, and CloudFormation provisions infrastructure rather than managing backups.

A business states it can tolerate at most 15 minutes of data loss during a disaster. Which metric does this define?

• a.RTO (recovery time objective)
• b.SLA
• c.RPO measured in dollars✓
• d.RPO (recovery point objective)

The recovery point objective is the maximum acceptable amount of data loss measured in time, so 15 minutes is an RPO. RTO defines acceptable downtime, an SLA is a service commitment, and RPO is measured in time, not dollars.