Amazon CloudWatch

A monitoring service to view the state and logs of AWS resources.

Back to glossary Go to related practice questions

In one sentence

A monitoring service to view the state and logs of AWS resources.

Overview

A monitoring service that manages metrics, logs, and alarms for AWS resources. Used to check Lambda logs and CloudFront metrics.

Common use cases

Reviewing Lambda error logs
Monitoring metrics for API Gateway and DynamoDB
Alarm notifications

CLF-C02 exam points

Handles metrics and logs
Sends threshold notifications via alarms
Used for operational monitoring

SAA-C03 exam points

Set retention periods for CloudWatch Logs
Trigger Auto Scaling based on metrics
Design anomaly detection and alarm notifications

Practical Use

Amazon CloudWatch is a monitoring service that integrates metrics, logs, and alarms for AWS resources and applications. Standard metrics such as EC2 CPU utilization, RDS connection counts, and Lambda execution times are collected automatically and can be used for dashboards and alarms without additional configuration. CloudWatch Logs is a service that collects and stores application logs, allowing centralized management of Lambda standard output, application logs on EC2, VPC Flow Logs, and more. CloudWatch Alarms send notifications via Amazon SNS to email, Slack, PagerDuty, and similar destinations when metrics exceed configured thresholds. Alarms can also trigger EC2 Auto Scaling to automatically scale instances up or down. Dashboards visualize metrics across multiple resources, and combining with Amazon EventBridge enables event-driven operational automation. On the cost side, custom metrics are charged based on the number of metrics and API requests. When ingesting large numbers of custom metrics, cost estimation is required in advance. CLF-C02 asks about CloudWatch as "the service that collects metrics and sends alarm notifications". A common topic is the three-service comparison among CloudTrail (API operation logs) and AWS Config (recording resource configuration changes and rule compliance). Clearly differentiating each service's role is important. Starting at SAA-C03, advanced usage patterns are tested — including query analysis with CloudWatch Logs Insights, using metric filters to generate custom metrics from logs, and top-N analysis via Contributor Insights.

Official Documentation

Cost notes

Charged based on log ingestion and storage volume
For MVP, set a short log retention period

Security notes

Do not log the full body of emails or contact form submissions
Limit log viewing permissions

Related services

AWS Lambda

Runs code only when needed, without launching servers.

Amazon API Gateway

The API entry point that invokes Lambda from the frontend.

AWS CloudTrail

An audit log service that records operation history on AWS.

AWS Config

A service to confirm whether AWS resource configurations conform to rules.

Related architecture diagrams

SERVERLESS / API / BASIC