NetApp Bets On Peak:AIO Acquisition To Provide New Performance Firepower In AI Data Battle

NetApp’s planned Peak:AIO acquisition strengthens its AI infrastructure and high-performance computing push, adding parallel NFS technology that channel partners can use to address enterprise AI factories, massive data workloads and tougher performance demands.

Intelligent data infrastructure technology developer NetApp Friday unveiled the planned acquisition of Peak:AIO, a developer of next-generation metadata architecture and high-performance parallel file systems.

Arindam Banerjee, NetApp’s chief platform and technology officer, told CRN that NetApp is acquiring Peak:AIO as a way to provide the orchestration to build large-scale environments.

“Peak:AIO works with a number of large national laboratories,” he said. “It is a part of Lattice, which we are very interested in.”

Lattice, which was unveiled in June, was developed in a collaboration involving Peak:AIO and the Los Alamos National Laboratory. Peak:AIO called Lattice the industry’s first open-source pNFS, or parallel NFS, metadata server.

Peak:AIO said Lattice provides for a new distributed metadata architecture aimed at eliminating one of the most persistent constraints in large-scale AI and high-performance computing infrastructure to provide ultra-fast, parallel access to massive datasets continuously and reliably to meet the needs high-performance AI workloads including training, running inference at scale, and serving agentic AI applications.

Banerjee said the Peak:AIO technology will help NetApp better address enterprise and institution requirements for building AI factories. The Peak:AIO technology is already in use at several national laboratories and high-performance computing environments, he said.

Banerjee declined to disclose how much NetApp is paying for Peak:AIO or when the acquisition is expected to close. More details around the financial arrangements and integration will be available at the close, he said.

Peak:AIO CEO Roger Cummings wrote in a LinkedIn post that the fit between NetApp and Peak:AIO is a compelling one for customers.

“NetApp brings decades of enterprise experience, a global customer base, and a reputation built on helping some of the world’s largest organizations manage their most critical data,” Cummings wrote. “Peak:AIO brings a different piece of the equation: the performance, scale and efficiency required as AI and HPC fundamentally change what customers expect from their infrastructure. Together, we have an opportunity to help customers build the reliability and enterprise strength they expect from NetApp, with the performance AI and HPC workloads demand.”

The acquisition of Peak:AIO is a great move for NetApp, said Ned Engelke, chief technology officer at Evotek, a San Diego-based solution provider and NetApp channel partner.

Peak:AIO’s parallel NFS client is something NetApp has not had in its portfolio, Engelke told CRN.

“It allows you to have more throughput bandwidth, and probably lower latency, for file workloads, which would be beneficial there,” he said. “And that had been a point of competition against NetApp, generally, in the space. So I like that.”

Engelke said there are three elements of infrastructure: how to find and manage data, how to move that data around, and how to get it to whatever application needs it.

“So it’s fun for me to see NetApp doing this,” he said. “NetApp has the first two pieces. They have a good, robust control plane and data plane, and so it’s nice to see them making moves around improving how to get data into whatever application is being served.”

For example, Peak:AIO could help improve the performance of AI, Engelke said.

“We want to go fast,” he said. “We’re running on a GPU. We’re going to do all these different things. NetApp has done a lot over the last 10 years or so to work well with high-performance computing environments. Before this acquisition, they did co-development work with Nvidia to feed the cache, things like that. They’ve been doing really good stuff, but they’re making acquisitions and moves to expand the part of how they interact with a customer that takes data from their whole data control plane and then starts to make it more available to applications. That’s why I’m excited about it.”