New Resource Optimization Recommendations Help Control AWS Costs

Amazon Web Services Inc. (AWS) announced a new custom resource optimization recommendation service for Amazon Elastic Compute Cloud (Amazon EC2) users.

To help cloud computing users control costs, the service analyzes:

  • Historical EC2 resource usage
  • Amazon CloudWatch metrics
  • A user's existing reservation footprint
This analysis is used to calculate ideal configurations and generate optimal recommendations that can be used as part of the AWS Cost Management service, where users can see advice for actions such as terminating unused EC2 instances and rightsizing under-used instances.

AWS said a new Recommendations summary page will show:

  • The number of optimization opportunities: The number of recommendations available based on past usage.
  • Estimated monthly savings: The projected monthly savings associated with each of the recommendations provided.
  • Estimated savings (percent): The approximate savings relative to a user's current instance costs, based on the recommendation list. This calculation is based on the On-Demand Instance price, so it may fluctuate from the prediction.

"For example," the company said in an announcement, "if your m5.2xlarge has a maximum utilization of 20 percent over the last 14 days, AWS may recommend downsizing that instance to m5.xlarge or m5.large and show you how much you can save based on your usage and your applicable m5 family reservations."

In a July 23 blog post, AWS advised potential users to note the following about the new service:

  • Instance support: Recommendations supports all major EC2 instance families (C, M, T, R, X) but does not currently support GPU-based instances.
  • Reserved Instances: Recommendations takes size-flexible Reserved Instance logic into account when generating recommendations. This might create recommendations to down-size instances for $0 in direct savings, which ultimately frees Reserved Instance hours you can use to cover other underutilized resources.
  • Opt-in process: After you opt into Resource Optimization recommendations, it can take around 24 hours for your recommendations to be generated. The opt-in process happens at the master account level.
  • Access controls: The payer account controls which linked account can access Cost Explorer and Recommendations from the account settings page. IAM users with "ViewBilling" permissions can view these recommendations.

"Resource optimization recommendations identify all available cost savings and rightsizing opportunities in your account or, for payer accounts, across your whole organization," AWS said. "Opportunities are identified even if they result in zero-dollar savings, as they might also improve your Reserved Instance coverage. Using these recommendations, you can enable members of your organization to easily identify and act on opportunities for improving cost efficiency."

Further information is available in the "Optimizing your Costs with Rightsizing Recommendations" user guide.

About the Author

David Ramel is an editor and writer for Converge360.


Subscribe on YouTube