Custom Metrics

5 minutes 5 Questions

Custom Metrics is an Auto Scaling feature that enables you to create and use your own application-specific metrics to drive the scaling decisions of your Auto Scaling groups. By using the CloudWatch API or the CloudWatch agent, you can collect, store, and analyze custom metrics that are relevant to…

Guide: Custom Metrics in AWS Solution Architect: Auto Scaling

What is it?
In Amazon Web Services (AWS), custom metrics are user-defined measures that are published to Amazon CloudWatch (a part of AWS auto scaling). These custom metrics give additional insights about the behavior of your applications, beyond standard metrics provided by AWS.
Why is it important?
Monitoring the performance of your applications is crucial, and using custom metrics allows you to create actionable alarms based on these metrics, or view more detailed statistics about your application resources. Custom metrics can monitor any aspect of your services which standard metrics may not cover, giving a holistic view of your services and allowing reactive and proactive responses to system behavior.
How does it work?
Custom metrics work by sending data to CloudWatch, which AWS will then monitor in real time. The frequency of your metrics depends on your needs, it could be every minute or less frequently. You can also store statistical data, set alarms or even automate actions based on custom metrics.
Exam Tips: Answering Questions on Custom Metrics
Understanding where and when to use custom metrics is key. Questions may ask about the functionality and benefit derived from using these metrics. A good tip is to revisit the AWS documentation and understand the theoretical and practical application of custom metrics. Remember, knowing the benefits of custom metrics, as in they fill in gaps standard metrics may not cover, will help in answering exam questions effectively.

Test mode:

Exam (Timed)

Practice (With explanations)

Start practice test

AWS Certified Solutions Architect - Custom Metrics Example Questions

Test your knowledge of Custom Metrics

Question 1

Your application runs on an EC2 instance and makes outbound network requests. You need to receive a CloudWatch alarm if the application's request latency exceeds a defined threshold. There is no existing CloudWatch metric that captures this per-request latency. What is the best way to implement this using CloudWatch?

Use the built-in EC2 instance network latency metric provided by Amazon CloudWatch. Publish a custom CloudWatch metric with the application's network request latency using PutMetricData, and create a CloudWatch alarm on that metric. Create a CloudWatch rule to raise an alarm when network latency is above the threshold without using a Custom Metric. Create an AWS Lambda function that triggers an alarm based on network latency only when it is higher than the specified threshold.

Correct Answer: Publish a custom CloudWatch metric with the application's network request latency using PutMetricData, and create a CloudWatch alarm on that metric.

EC2 does not provide a built-in metric for per-request network latency. The correct approach is to instrument the application to record request latency and publish those values as a custom CloudWatch metric using the PutMetricData API (or an equivalent method such as EMF). You can then create a CloudWatch alarm on that metric to alert when latency exceeds the threshold. A CloudWatch rule/EventBridge rule cannot measure latency or trigger an alarm without a metric, and using a Lambda function to detect latency and "trigger" an alarm is unnecessary, adds operational complexity, and is not the intended pattern for CloudWatch alarms.

Question 2

A company wants to monitor a custom application's cache hit and cache miss rates using Amazon Cloudwatch. Which Custom Metric should be used to create a monitoring solution?

Calculate the cache hit and cache miss rates using ELB access logs and then create a CloudWatch alarm. Publish the application's cache hit and cache miss rates using the PutMetricData API to create a Custom Metric in CloudWatch. Use an AWS Lambda function to calculate cache hit and cache miss rates based on the default CloudWatch Metrics. Use AWS WAF to monitor the cache hit and cache miss rates based on HTTP request headers.

Correct Answer: Publish the application's cache hit and cache miss rates using the PutMetricData API to create a Custom Metric in CloudWatch.

The correct answer is to publish the application's cache hit and cache miss rates using the PutMetricData API to create a Custom Metric in CloudWatch. This allows monitoring the custom application's cache hit and miss rates in real-time. The other options do not address creating a Custom Metric or are not applicable to the scenario.

Question 3

A mobile app hosted on an EC2 instance requires monitoring the number of users that are currently active. Which of the following Custom Metrics in CloudWatch will help achieve this requirement?

Use the default CloudWatch Metrics for the EC2 instance, as it already includes the active users count. Create a Custom Metric using a script on the EC2 instance to send the current active users count to CloudWatch. Create an Amazon S3 bucket and store the active users count every minute, then use this data as a Custom Metric. Monitor the active users count using an Amazon RDS database and send it to CloudWatch.

Correct Answer: Create a Custom Metric using a script on the EC2 instance to send the current active users count to CloudWatch.

The correct option is to create a Custom Metric using a script on the EC2 instance to send the current active user count to CloudWatch, as it directly addresses the need to count active users in real-time. The other options either do not address active users count or do not involve creating a Custom Metric in CloudWatch.

🎓 Unlock Premium Access

AWS Certified Solutions Architect - Associate + ALL Certifications

🎓 Access to ALL Certifications: Study for any certification on our platform with one subscription
5645 Superior-grade AWS Certified Solutions Architect - Associate practice questions
Unlimited practice tests across all certifications
Detailed explanations for every question
AWS Certified Solutions Architect: 5 full exams plus all other certification exams
100% Satisfaction Guaranteed: Full refund if unsatisfied
Risk-Free: 7-day free trial with all premium features!

Start Your Free 7-Day Trial