How to Prevent AWS Cost Traps
3 Major Cloud Cost Management Challenges
Cost management is typically the first issue people want to tackle when they enter the cloud. However, contrary to popular belief, cloud cost optimization is not a one-time fix. The most effective cloud cost management requires vigilance and a commitment to continual review as things change quickly and constantly with AWS. After our post on Amazon Web Services (AWS) cost issues, we’ve found that these cost traps arise more and more often.
CloudCheckr can automate the detection of each of the cloud cost management issues listed below, while also informing you of their value and corresponding savings.
Here are three additional AWS cost management challenges, explanations of why they are important, and recommendations on how to mitigate the issues.
1. EC2 Reserved Instance (RI) Possible Mismatch
Checks all On-Demand instances to see if any may have been intended to utilize Reserved Instance pricing.
Why it’s important:
Reserved instances offer a greatly reduced hourly rate often in exchange for a one-time, up-front fee as well as a one or three year commitment. Once a reserved instance is purchased, a user must launch an On-Demand EC2 instance that matches all of the criteria of the Reserved purchase to be billed at the reduced price. This criteria can include the region of the instance, as well as the instance type, size, and operating system.
If you do not properly match all of the criteria, your instance will not use the lower Reserved Instance pricing. It is very difficult to detect this type of mistake in the AWS Management Console, and can go unnoticed.
CloudCheckr’s Best Practice Check locates any On-Demand instances that could take advantage of any unused Reserved Instances and highlights those instances. If the unmatched criteria is the Availability Zone, AWS allows you to change the Availability Zone of the reserved instance.
2. EC2 On-Demand Instances Not Using Reserved Instance (RI) Pricing
Checks the number of RIs being utilized against the number of On-Demand instances running. Lists the number of On-Demand instances that are not being properly billed as RIs.
Why it’s important:
RIs offer a greatly reduced hourly rate in exchange for an up-front commitment. Once an RI is purchased, a user must launch an On-Demand EC2 instance that matches all of the criteria of the RI purchase to be billed at the reduced price. This criteria includes the region of the instance, platform, type, and AZ.
All On-Demand instances should be reviewed regularly to see if they should be converted to RIs in order to take advantage of lower pricing. All instances that are stable and will be in use for at least six months are good candidates to be upgraded to Reserved.
3. Unattached EBS Volumes
Checks for all EBS Volumes that are not attached to any EC2 instances.
Why it’s important:
Elastic Block Store (EBS) provides block level storage volumes for use with EC2 instances. These volumes can be attached to any EC2 instances within the same AZ. While an EBS volume can only be attached to one instance at a time, an instance may have multiple volumes attached to it.
EBS volumes also carry on independently from the life of any instance they are attached to. Users can take snapshots of their EBS volumes to act as backup, or to be used as a baseline for new volumes. These EBS snapshots are incremental, meaning that only the blocks on the volume that have changed since the previous snapshot will be saved. All snapshots are saved within S3, which means they are stored redundantly and in multiple Availability Zones.
EBS volumes incur costs based on their size, performance specifications and number of I/O requests made each month. If a volume is not being utilized, it will still incur charges for provisioned storage. EBS snapshots also incur fees based on the size of the data being stored in S3 Buckets.
All storage volumes that are not attached to any instances should be reviewed. Depending on the volume and what it is comprised of, a number of choices can be made.
Delete it if the volume is no longer needed or useful.
Take a snapshot, then delete it if the volume is unique and will either be needed in the future, or needed to create new volumes, a
Attach the volume to a running EC2 instance and copy its data to an S3 bucket if the volume contains data that needs to be retained, but the volume itself is not needed (storage in S3 is less expensive than storing snapshots). The volume can then be detached from the instance and deleted.
Saving Money on AWS
It may sound like constantly checking all of these things will take up a lot of your time, but on the contrary—cost management does not require tremendous effort if you utilize automation. Automating reports ensure your attention is only spent on high-value items that impact your results while avoiding the low-value, time-wasting items that barely make a dent in your cloud bill.
This identification and actionable insight is one of CloudCheckr’s specialities. CloudCheckr automates the detection of each of these issues. It will also provide each issue’s corresponding value and savings so you can turn off or downsize any EC2 instances costing you thousands per month. Automation like this ensures your team will no longer waste time tracking down minor issues, like the owners of idle EC2 instances costing you $20 per month.
See how CloudCheckr can drastically simplify your cloud cost management. Sign up for your free CloudCheckr trial now.
Only CloudCheckr delivers total visibility, making the most complex cloud infrastructures easier to manage. CloudCheckr customers deploy our total visibility platform to secure, manage, and govern the most sensitive environments in the world—from government agencies to large enterprises to MSPs. CloudCheckr industry-leading solutions include Cost Management, FinanceManager, Cloud Security, Total Compliance, Inventory & Utilization, and Cloud Automation.