Which AWS Service Monitors EC2 Disk Write Spikes?

Why the Answer Is Correct

Amazon CloudWatch is AWS's monitoring and observability service: it ingests time-series metrics for EC2 instances and lets you graph them, set thresholds and trigger alarms or Auto Scaling actions. Disk write activity is exactly that kind of performance telemetry, so a spike can be detected with a CloudWatch alarm on metrics such as disk write operations or write bytes. Because CloudWatch is the native metrics store for AWS resources, it is the only option here that provides real-time, quantitative disk-level monitoring. The community agreed: as Yomijohnson wrote, "It provides metrics, including those related to disk performance", and its_tima described CloudWatch as the service for "monitoring and observing resources in real-time".

Why the Other Options Are Wrong

AWS CloudTrail (A) is an audit and governance service — it records API calls and account activity (who did what, when, from where), not CPU, disk or network performance counters, so it cannot detect a write spike. AWS Health Dashboard (B) reports service health events, scheduled changes and issues affecting your account or specific AWS services; it is about availability notifications, not EC2 disk throughput. AWS Trusted Advisor (C) inspects your account against best-practice checks (cost, security, fault tolerance, service quotas) and offers recommendations, but it does not stream per-instance disk metrics or fire real-time alarms on write volume. Each of these watches a different layer of the stack, which is why only CloudWatch answers a performance-monitoring question.

Community Comment Notes

All three comments select D, and the reasoning is consistent: monitoring resource metrics in near real time. Yomijohnson emphasises that alarms can be configured "to notify you of potential disk write spikes or other performance issues", which is precisely the use case in the question. its_tima frames CloudWatch as the tool that gives "data and actionable insights" for application monitoring and resource optimization. d00b229 simply records D without elaboration. One nuance worth remembering for the exam: instance-level disk metrics for EC2 come from the CloudWatch agent, while basic metrics like CPUUtilization are available by default.