A multinational enterprise manages its development environments across 50 separate AWS accounts consolidated under a single organization in AWS Organizations. The cloud engineering team is tasked with implementing a cost management solution that will automatically detect unexpected, erratic spikes in daily data transfer fees and instance usage within 24 hours of occurrence. Due to the highly variable workloads in each developer account, the solution must use machine learning to establish a dynamic baseline and adaptively alert on spend anomalies rather than relying on fixed, static thresholds. Which approach will satisfy these requirements with the lowest administrative effort?
- Create an AWS Cost Anomaly Detection monitor for all member accounts within the organization, define an alert subscription with a threshold based on the daily dollar impact, and configure notifications to be sent to an Amazon Simple Notification Service (Amazon SNS) topic.Answer
- BEstablish daily cost budgets in AWS Budgets for each account, set the budgeted amount to match the average historical spend of the previous month, and configure an email notification to trigger when actual costs exceed 110% of the budgeted amount.
- CCreate a custom Python script running on a daily Amazon EventBridge cron schedule that calls the AWS Cost Explorer API to query the previous day's usage, performs statistical standard deviation checks on the retrieved data, and sends alerts to an Amazon SNS topic.
- DConfigure AWS Billing Conductor to define custom billing groups for each development account, set up billing rules to filter out standard usage patterns, and configure a CloudWatch alarm to trigger when the pro forma billing data deviates from the baseline.
Answer
Create an AWS Cost Anomaly Detection monitor for all member accounts within the organization, define an alert subscription with a threshold based on the daily dollar impact, and configure notifications to be sent to an Amazon Simple Notification Service (Amazon SNS) topic.
The correct option is the one proposing AWS Cost Anomaly Detection. AWS Cost Anomaly Detection is a fully managed service that uses machine learning to continuously monitor cost and usage, detect unusual spend patterns, and minimize false alerts by adapting to historical usage trends. It natively integrates with AWS Organizations to monitor all member accounts, and notifications can be routed via Amazon SNS within 24 hours of anomaly detection with zero custom scripting.
Step-by-Step Solution
Key Concept
AWS Cost Anomaly Detection leverages machine learning to dynamically baseline usage and alert on spend anomalies with low operational overhead.
Estimated Time:2m 0s