A company has deployed a serverless application utilizing Amazon DynamoDB in a production AWS account. The workload has highly variable but predictable weekly patterns, causing daily costs to fluctuate. A SysOps administrator needs to configure a monitoring solution that uses machine learning to detect unexpected cost spikes in DynamoDB usage while minimizing false alerts from normal weekly variations. When an anomaly is detected, the administrator must receive an immediate notification via email and SMS.
Which solution meets these requirements with the least administrative effort?
- ACreate an AWS Budget with a daily budget threshold based on a 3-month historical average. Configure the budget to send notifications to an Amazon SNS topic.
- BConfigure an Amazon EventBridge rule that monitors CloudWatch DynamoDB metrics. Trigger an AWS Systems Manager Automation runbook that evaluates the metric value and publishes an alert to Amazon SNS.
- Create an AWS Cost Anomaly Monitor using the AWS Service monitor type for Amazon DynamoDB. Create an anomaly subscription with an Amazon SNS topic as the alerting channel, and subscribe the email and SMS endpoints to the SNS topic.Cevap
- DApply a custom billing tag to the DynamoDB tables, activate the tag in the AWS Billing Console, and configure an AWS Budget using the activated tag as a filter to alert when costs exceed the standard forecast.
Cevap
Create an AWS Cost Anomaly Monitor using the AWS Service monitor type for Amazon DynamoDB. Create an anomaly subscription with an Amazon SNS topic as the alerting channel, and subscribe the email and SMS endpoints to the SNS topic.
The correct solution involves creating an AWS Cost Anomaly Monitor of the AWS Service type to dynamically track Amazon DynamoDB costs using machine learning to account for seasonal variations. An anomaly subscription linked to an Amazon SNS topic allows direct integration for sending email and SMS notifications without relying on static thresholds.
Adım Adım Çözüm
Anahtar Kavram
AWS Cost Anomaly Detection is a machine-learning-driven service that baselines cost patterns to detect unexpected spend spikes, which is more effective than static budgets for variable workloads with seasonal trends.
Tahmini Süre:1m 30s