Last month, the cloud management platform provider nOps Inc. joined theCUBE to discuss leveraging artificial intelligence to optimize cloud costs in a pay-for-savings only model.

The company is always looking for ways to optimize customer spend, according to JT Giri (pictured), chief executive officer of nOps. What that has led to is a realization that a lot of workloads are moving to containers and particularly to Kubernetes. That’s where nOps Karpenter Solution (nKS) comes into play, further trying to optimize customer spend.

“Everything we do on nOps, we focus on automation, where a customer doesn’t have to go look at a report and take action,” Giri said. “nKS automatically looks at the spot pricing, it looks at your reserve instances, commitments, it looks at your savings plans and it picks the right nodes to save you the most amount of money.”

Giri spoke with theCUBE industry analyst Lisa Martin at the “Analytics and Cost Optimization” AWS Startup Showcase, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed the company’s use of artificial intelligence in nKS to automate and optimize cloud costs for organizations. (* Disclosure below.)

Reducing Elastic Kubernetes Service costs

So how does nKS help organizations simplify its EKS management and optimize its AWS costs? Most customers are using EKS to run their Kubernetes environment, according to Giri.

“A cluster autoscaler is kind of constrained by node groups. Whenever you are scaling up, your infrastructure is constrained by instance types,” Giri said. “When it’s launching new pods, essentially it’s first launching the infrastructure. Then it’s retrofitting your application to that infrastructure.”

AWS launched its open-source solution Karpenter in 2021, which frees users from node groups and autoscaling constraints, and nKS takes that a step further, according to Giri.

“We make sure that Karpenter is aware of your reserve instances reservations and is aware of your savings plan reservations. We also reconsider pricing, real time, to make sure we pick the best optimal cost, optimal instances types to save you the most amount of money,” Giri said. “In many cases, we can reduce your EKS cost by 50% without actually requiring a lot of engineering work.”

Gaps with existing EKS autoscalers

Many customers running their Kubernetes workload using EKS are moving to Karpenter, which is much better positioned to handle EKS workloads given it is closer to the Kubernetes API and is provisioning resources based on application requirements, according to Giri. That’s where nOps saw an opportunity when it came to customer migration.

“Let’s say if they were running workloads on Spot, if they have savings plan commitments they made or reserve instances, they were actually losing money,” Giri said. “We saw an opportunity where we could do this automatically and pick the best cost-optimal node types so customers don’t have to think about it.”

That’s because the company believes that engineers should focus on innovation, not spending their time on learning AWS pricing plans, according to Giri.

“We build solutions and bring solutions to the market that free up engineers,” he said.

nOps is part of the expanding FinOps movement, an area of focus among firms that specialize in providing visibility and accountability in cloud spending throughout the enterprise. Having recently launched nKS, the company has recently onboarded a number of customers, according to Giri.

“There’s a lot of interest, because a lot of customers are moving to Karpenter. We also realize customers spend a lot of time just managing Kubernetes and Karpenter configurations,” he said.

nOps is not a point solution; it provides a complete FinOps platform starting with providing visibility, according to Giri.

“We have showbacks and chargeback type of functionality, and we give that functionality for free,” he said. “Then we look at resources that could be paused off-hours, like your dev environments, your sandbox, staging. Then we provide a one-click solution that your developers could use to pause these resources.”

That’s where nKS comes into play if resources are running on Kubernetes, especially around EKS, according to Giri.

“We provide autopilot configuration management for Karpenter, where we can pick the right note type, so you can save the most amount of money,” he said. “In the backbone, we’re looking at your commitments, we’re looking at your savings plan commitments and reserve instances, and we’re adjusting those accordingly.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the “Analytics and Cost Optimization” AWS Startup Showcase event:

(* Disclosure: nOps Inc. sponsored this segment of theCUBE. Neither nOps nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE