3 Tactics This Company Used to Reduce Its AWS Cloud Costs By 48%
Just like many things in life, the cloud is what you make of it. If you conduct a thorough analysis of your performance profile and use that data to make strategic cloud decisions, you’ll get to enjoy maximum performance at the lowest possible cost. But if you approach your cloud management without an active management approach, you’re almost guaranteed to suffer from cloud sprawl and overspend.
Even organizations that right-size their cloud configuration upon their move to the cloud often fail to manage their cloud effectively after the migration. A Cloudamize industry report recently published the results of a comprehensive analysis of over 10,000 machines and found that 33% of the instances on AWS are over-provisioned and that 33% and 19% of the instances on AWS and Azure (respectively) are idle. What do these findings mean? They mean that, despite best intentions, most companies are paying for more than they need or paying for what they’re not even using. These organizations could benefit from active cloud cost management and automated data analysis.
A North American trade compliance organization turned to Cloudamize to get its cloud costs under control. At the time, costs were double its budget. It wanted to reduce its cloud costs but keep its performance level intact before moving the rest of its infrastructure to AWS. Thanks to our automated data analysis platform, this customer was able to cut cloud costs without compromising their performance in these 3 ways:
1. Identify and Eliminate Waste:
Instead of purchasing on-premises capacity to make sure that they’re covered for occasional peaks in demand, the cloud allows customers to stop paying for compute and storage resources they don’t need when they’re not using them—but there’s a catch. Companies can only identify and eliminate waste if they’re actively managing performance, usage, and costs, and many companies fail to do so effectively. As an example, companies frequently make the mistake of turning off compute resources when they aren’t needed but they forget to reflect that change in their storage volume.
With better insight into idle compute resources and unused storage powered by automated data analysis, Cloudamize helped this organization identify and eliminate waste—to the tune of $332,000 on idle resources and $11,300 in unused storage annually. Critically, these were resources that this customer could eliminate without seeing a negative impact on performance.
2. Purchase Reserved Instances:
While on-demand pricing for cloud instances can give organizations the flexibility to adapt to changes in your demand, that flexibility comes at a high cost. In fact, on-demand pricing is actually the most expensive pricing model and most companies could control their cloud costs more effectively by switching to purchasing reserved instances (RIs) in advance.
Based on past and current performance and usage analysis, Cloudamize helped this company to both plan current capacity and to forecast future compute, storage, and network capacity needs. An automated data analysis recommended that the company reduce its on-demand hours from over 1 million to less than 40,000 and more accurate advance RI purchase planning helped the customer significantly reduce spending.
3. Right-Size Instances:
Another common mistake that can lead to unwanted cloud costs is failing to continually right-size instances. When a workload’s performance requirements or profile changes, switching to a different instance can help companies avoid over- or under-provisioning. Additionally, since cloud providers introduce new instances and change pricing from time to time, companies may also be able to save costs by switching workloads to an instance that has just been introduced or recently reduced in price.
Identifying your right-size compute resources is a gargantuan task, which is why automated data analysis is so important. We enabled this customer to achieve savings of over $70,000 annually by right-sizing instances that were over- or under-provisioned.
One of the biggest benefits of the cloud is that companies who migrate their infrastructure only need to pay for what they’re actually using. But at the end of the day, this company was only able to control cloud costs by actively managing their cloud—and accurately managing its cloud costs was only been possible with automated performance and usage analysis. All told, the end result of these analyses for this customer resulted in a 48% AWS cost reduction.