DDN targets GPU efficiency with AI data infrastructure as the make-or-break layer

The race to build AI factories is well underway, and the winning organizations have learned that AI data infrastructure determines whether or not GPU investments pay off, while others are still scrambling to assemble workable solutions.

That divide is the clearest indicator from the field, said Alex Bouzari (pictured), chairman, co-founder and chief executive officer of DataDirect Networks Inc. DDN sits at the core of some of the world’s largest AI deployments, including hundreds of thousands of GPUs for xAI. Bouzari said the pattern is consistent. GPU utilization rates distinguish organizations where AI delivers measurable financial outcomes from those where expensive infrastructure is underused.

“The world is bifurcating into those enterprises, those nations where the GPUs are being fully utilized at highest levels of efficiency and the ones where the GPUs are sitting idle,” Bouzari said. “The ones who are not getting it right are trying to cobble solutions together, and by doing so, they’re spending a lot of money, but it’s wasted capital.”

Bouzari spoke with theCUBE’s John Furrier at the RAISE Summit during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how AI data infrastructure has become the defining layer of the AI stack, why sovereignty is reshaping deployment architectures and what distributed edge computing means for the next phase of AI factory design. (* Disclosure below.)

AI data infrastructure and sovereignty converge as nations build their own AI factories

One of the dominant themes at RAISE, Bouzari said, is data sovereignty. DDN is currently involved in a dozen sovereign AI projects, and the message from governments is consistent: they want access to frontier AI capabilities without their data crossing borders. This demand is pushing deployments of a new class of nationally scoped AI data infrastructure that goes beyond enterprise IT.

“Every country really needs to put in place AI infrastructures, AI factories that are sovereign in nature,” Bouzari said. “Give us the independence where we know that our data stays in our country and does not trickle out to some other place.”

In terms of scale, Bouzari pointed to Salesforce as an example of what optimized AI data infrastructure delivers, a 70% increase in GPU productivity after DDN was deployed. He said NVIDIA’s own internal use of DDN for the past eight years is the most credible validation in the market, citing Jensen Huang’s public statement that NVIDIA supercomputers would not be possible without DDN.

“DDN is for data what NVIDIA is to compute,” Bouzari said. “The combination of the two delivers the end-to-end SLA, which is really all that matters.”

The next architectural challenge involves stitching together a globally distributed hierarchy of AI nodes. This includes large factories in the 25 to 100 megawatt range handling model training, connected to edge data centers, collecting real-time data from autonomous vehicles, robots and sensors across the world. DDN’s distributed data platform, Infinidat, was designed eight years ago with such a distributed model in mind, and Bouzari said the complexity of gathering those layers into a coherent AI data infrastructure pipeline is only expanding as agentic workloads scale.

“Connecting multiple edge data centers, globally distributed, into a small number of monolithic data centers, into a multi-cloud environment — that is a very difficult problem to solve,” Bouzari said. “But eight years ago when we started the development of our Infinidat technology, that’s how we looked at it, because NVIDIA at the time told us this is what it’s going to look like.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of RAISE Summit:

(* Disclosure: TheCUBE is a paid media partner for the RAISE Summit event. Neither Solidigm, the headline sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE

A message from John Furrier, co-founder of SiliconANGLE:

Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.

15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more

11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.

About SiliconANGLE Media