Monitoring a custom AWS service

Scenario

Sarah is a tenant administrator at Apex Global. The company uses AWS services in their day-day-day operations. Sarah lately feels that the response latency of the AWS Polly service has increased. She wants to monitor the response latency and rule out network issues. Sarah wants to be notified of changes in the response latency so she can take an action on time. 

Apex Global uses BMC Helix Operations Management as their monitoring system. Sarah knows that she can use the Amazon Web Services KM to monitor their AWS services. However, Sarah realizes that the AWS Polly Response Latency metrics are not a part of the Amazon Web Services KM. How can Sarah set up an alert to understand changes in the AWS Polly response latency?

Benefits

The good news is that AWS KM has the capability to extend the monitoring and introduce new metrics dynamically by using Custom Metrics. For the above example of the AWS Polly Metrics, we can simply use specify the Namespace Name as “AWS/Polly” and Dimension Name as “Instance,MetricClass,Operation.” So just by doing this configuration AWS KM will generate corresponding Monitoring Type and start monitoring the newly configured attributes.

Workflow

Results

Apex Global can now:

• Monitor the following AWS services that they use:
  • Amazon CloudHSM Cluster
  • AWS Account
  • AWS Redis Cache
• View the alarms generated about the performance of these services.