Cloudwatch Metrics

Cloudwatch metrics are criteria or dimension of an AWS service that is monitored and can be displayed in graphical or numerical format. For example, the CPU usage of EC2 instances or RDS instances or Lambda invocations or API Gateway calls can be monitored and graphed.

In AWS, there are many metrics that is available by default. However, certain metrics, for example, memory usage of EC2 instances are not available - the reason being AWS monitors to the VM host but not the actual EC2 instances. However, custom metrics can be added to Cloudwatch.

To view the metrics, log into AWS console, navigate to Cloudwatch service and click on Metrics from the left menu. All the available metrics are displayed group by services.

To add metrics to the graphs, simply click on the service and then add the required metric. For example, in this demo, the CPU utilization and queries metrics are added and the following graph is displayed. As expected, the CPU usage is proportional to the queries.