Migrating to the cloud brings huge benefits to any company. But cloud usage comes at a cost, and optimizing spend to get maximum performance is a constant balancing act.

So when Michael D’Aniello (pictured), platform architect for VMware Carbon Black at VMware Inc., cut 30% from cloud spend by migrating from Intel to AMD-based EC2 instances, he wanted to share the secret.

“We’re leveraging a lot of our instances to run our EKS clusters, which is a managed service from AWS to run our Kubernetes clusters, and we identify that we can take a bunch of those instances and gain some cost optimization benefits by selecting from Intel to AMD processors,” D’Aniello said. “Initially, we had measured out roughly a 10% reduction in cost just for selecting that instance type. But yeah, we actually learned we gained quite a bit more.”

Speaking with John Furrier, host of theCUBE, SiliconANGLE Media’s livestreaming studio, during AWS re:Invent, D’Aniello discussed VMware Carbon Black’s AMD-based instance migration. (* Disclosure below.)

A simple change brings cost savings and performance upgrade

As a platform architect, D’Aniello is always looking for ways to keep the infrastructure underlying the Carbon Black cloud on the cutting edge. A key aspect of the job is not only keeping an eye on new technologies that are coming out but “trying to get the most bang for your buck while you’re at it,” he said.

It was this quest for cost optimization that led D’Aniello to investigate AMD-powered EC2 instances. Carbon Black was leveraging a lot of Amazon Web Services Inc.’s Elastic Compute Cloud (or EC2) instances to run its Amazon Elastic Kubernetes Service (EKS) clusters, and D’Aniello discovered that he could save money and increase performance by switching those instances from Intel to AMD processors.

“Upfront, the instance itself is 10% cheaper. However, we found out that we had to run far less instances because of that performance increase. So we ended up saving roughly 30%,” D’Aniello said.

Even with the cost savings and performance uptick, D’Aniello might not have opted to trial the AMD processors if the migration had involved complex code changes. But, the change was minor.

“We had to make just a single line change to change that instance type in our config and then roll that out across our regions.”

The team uses the infrastructure-as-code tool Terraform, which AWS Partner HashiCorp Inc. provides as a layer for automated provisioning. This enabled the migration to happen without the developers having to make any application changes, according to D’Aniello. In fact, the only change needed was a one-letter addition to the instance identifier.

“It’s as easy as just the A after the name,” D’Aniello said, explaining how Carbon Black differentiates Intel from AMD instances by the addition of the letter A. For example, if the Intel instance was “C5.4XL,” the AMD one will be “C5A.4XL.”

“The greatest part about the story from my perspective is the ease to migrate over and to switch to these instance types, and then you just immediately gain that cost optimization benefit,” D’Aniello stated.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of AWS re:Invent. (* Disclosure: Advanced Micro Devices Inc. sponsored this segment of theCUBE. Neither AMD nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE