A SysOps Administrator is monitoring a latency-sensitive application hosted on Amazon EC2 instances. The application publishes a custom metric named RequestLatency to Amazon CloudWatch. The administrator needs to design an automated remediation workflow that triggers within 30 seconds when average latency exceeds 500 milliseconds. Which TWO actions should the SysOps Administrator take to implement this monitoring and remediation capability? (Select TWO.)
- Configure the application to publish the RequestLatency metric with the StorageResolution parameter set to 1.Answer
- Create a CloudWatch alarm for RequestLatency with a Period of 10 seconds and Evaluation Periods set to 3.Answer
- CEnable detailed monitoring on the EC2 instances to increase the publishing frequency of custom metrics.
- DCreate an Amazon EventBridge rule that triggers directly on the RequestLatency metric data injection events.
- EIncrease the Auto Scaling group cooldown period to allow custom metrics sufficient time to aggregate before the alarm is evaluated.
Answer
Configure the application to publish the RequestLatency metric with the StorageResolution parameter set to 1, and create a CloudWatch alarm for RequestLatency with a Period of 10 seconds and Evaluation Periods set to 3.
To detect latency issues and trigger a workflow within 30 seconds, a high-resolution alarm is necessary. CloudWatch supports high-resolution alarms with periods of 10 seconds or 30 seconds. To support this, the metric itself must be published as a high-resolution metric by specifying a storage resolution of 1 second (StorageResolution=1) during publication. An alarm period of 10 seconds with 3 evaluation periods provides a 30-second evaluation window to detect breaches.
Step-by-Step Solution
Key Concept
CloudWatch High-Resolution Metrics and Alarms
Estimated Time:2m 30s