Amazon CloudWatch のアイコン

Amazon CloudWatch Essential2009年〜

A service that provides monitoring, log management, and alarms for AWS resources and applications

About 3 min read

What It Does

Amazon CloudWatch collects, visualizes, and monitors metrics (numerical data), logs, and events from AWS resources and applications. It displays metrics like CPU utilization and memory usage on real-time dashboards and sends alarm notifications when thresholds are exceeded. With built-in log aggregation, search, and analysis capabilities, CloudWatch serves as a centralized operations monitoring platform for your AWS environment.

Use Cases

CloudWatch is used across all operational monitoring scenarios: auto-scaling EC2 instances when CPU utilization exceeds 80%, sending Slack notifications when Lambda function error rates spike, searching application logs for specific error patterns, and much more.

Everyday Analogy

Think of it like a car dashboard. The dashboard displays gauges for speed, fuel level, engine RPM, and coolant temperature, letting you check the car's status in real time. Warning lights come on when something is wrong. CloudWatch is the dashboard for your AWS environment, monitoring the status of each service and alerting you with alarms when problems arise.

What Is CloudWatch?

Amazon CloudWatch is the core operations monitoring service for AWS. Nearly all AWS services, including EC2, Lambda, RDS, and S3, automatically send metrics to CloudWatch. You can visualize these metrics in graphs, set thresholds to trigger alarms, and aggregate logs for search and analysis. It is an essential service for stable AWS environment operations.

Metrics and Alarms

Metrics are time-series numerical data representing the state of AWS resources. Examples include EC2's CPUUtilization and Lambda's Errors count. Alarms set thresholds on metrics and trigger actions (SNS notifications, Auto Scaling, Lambda execution, etc.) when conditions are met. For example, you can configure an alarm for "notify when CPU utilization exceeds 80% for 5 consecutive minutes."

CloudWatch Logs

CloudWatch Logs is a service for aggregating, storing, and searching application and system logs. Lambda function execution logs are automatically sent to CloudWatch Logs. For EC2 instance logs, install the CloudWatch Agent to send them. The Logs Insights feature lets you search and analyze logs at high speed using a SQL-like query language. For example, you can run queries like "show error logs from the past hour sorted by frequency." For practical know-how on CloudWatch Logs, you can also learn from specialized books on Amazon.

Getting Started

CloudWatch is available as soon as you create an AWS account. When you open the CloudWatch console, basic metrics from your AWS resources are already being collected. Create a dashboard, add metric graphs, and set up alarms to establish a basic monitoring setup. To send custom metrics (application-specific values), use the PutMetricData API from the AWS SDK.

Things to Watch Out For

  • Basic metrics are free, but detailed monitoring (1-minute intervals) and custom metrics incur additional charges. Consider whether standard monitoring (5-minute intervals) is sufficient to keep costs down
  • CloudWatch Logs retention is unlimited by default. Unnecessary log accumulation increases storage costs, so set appropriate retention periods for each log group
  • If alarm evaluation periods and thresholds are not properly configured, temporary spikes can cause frequent false alarms. Requiring consecutive threshold breaches across multiple periods helps reduce false positives
共有するXB!

Related Services

AWS CloudTrail iconAWS CloudTrailA service that records and monitors API activity in your AWS accountAmazon SNS iconAmazon SNSA fully managed Pub/Sub messaging and notification serviceAWS Lambda iconAWS LambdaA serverless compute service that runs code without server managementAWS Auto Scaling iconAWS Auto ScalingA service that automatically scales EC2, ECS, DynamoDB, and other resources based on demand

Related Articles

How AWS Keeps Time Internally - Amazon Time Sync Service and Leap Second Smearing DesignLearn how Amazon Time Sync Service works, how GPS and atomic clocks provide high-precision time sources, the design decision to absorb leap seconds through smearing, and why time synchronization matters in distributed systems.Implementing Feature Flags with AWS AppConfig - Safe Configuration Deployment and RollbackRoll out configuration changes independently from code deployments using Linear and Exponential strategies. Ensure safety with automatic rollback triggered by CloudWatch alarms.Why Auto Scaling Scales Out Fast but Scales In Cautiously - The Design Intent Behind Asymmetric Decision LogicThis article explains why EC2 Auto Scaling executes scale-out immediately while applying a cooldown period for scale-in, the flapping prevention mechanism, and the internal logic of target tracking scaling.Demand-Driven Infrastructure with AWS Auto Scaling - Designing and Optimizing Scaling PoliciesLearn how to use target tracking, predictive, and scheduled scaling policies effectively, and optimize costs with mixed instances policies that leverage Spot Instances.Lessons from AWS Incident Reports (COE) - How Past Major Outages Shaped Design PrinciplesAnalyze the root causes of past major incidents including the S3 outage, us-east-1 DNS failure, and Kinesis outage from AWS's published Correction of Errors (COE) and incident reports, and explain how they changed AWS's design principles.

More in This Category

AWS AppFabric iconAWS AppFabric NewA service that standardizes and aggregates audit logs from SaaS applicationsAWS AppConfig iconAWS AppConfig SpecializedA service for safely deploying and managing application configurationAWS Chatbot iconAWS Chatbot SpecializedAn interactive agent that delivers AWS notifications and enables resource operations through Slack and Microsoft TeamsAWS CloudFormation iconAWS CloudFormation EssentialAn IaC service that defines and provisions AWS resources as code using templatesAWS CloudTrail iconAWS CloudTrail EssentialA service that records and monitors API activity in your AWS accountAWS Config iconAWS Config PopularA service that records and evaluates AWS resource configuration changes and continuously monitors complianceAWS Control Tower iconAWS Control Tower PopularA service that automates multi-account environment setup and governanceAmazon CloudWatch RUM iconAmazon CloudWatch RUM NewA real user monitoring service that collects client-side performance and error data from web applications

Similar Articles and Services

Building Unified Monitoring with Amazon CloudWatch - Designing Metrics, Logs, and AlarmsBuild unified monitoring with the three pillars of metrics, logs, and alarms. This article covers interactive analysis with Logs Insights, high-precision notifications with composite alarms, and leveraging Embedded Metric Format.Operational Monitoring in Practice - Achieving Full-Stack Observability with CloudWatchLearn about operational monitoring design with AWS CloudWatch, including metrics collection, log analysis, and alarm configuration for comprehensive observability.Amazon CloudWatchA fully managed monitoring service that provides unified monitoring and analysis of metrics, logs, and events for AWS resources and applicationsWhy CloudWatch Logs Costs More Than You Expect - The Cost Structure of Ingestion, Storage, and AnalysisThis article explains the pricing structure where CloudWatch Logs ingestion costs over 20 times more than storage, the cost explosions caused by VPC Flow Logs and Lambda logs, and optimization strategies using the Infrequent Access class and S3 export.The Integration Power of the AWS Observability Stack - Operational Transparency Through CloudWatch, X-Ray, and CloudTrailThis article examines the integration level of the AWS observability stack centered on CloudWatch, X-Ray, and CloudTrail, comparing it with Azure Monitor and GCP Cloud Logging to explain how the three pillars of metrics, traces, and logs drive differences in operational quality.