The VMware Cloud Foundation Division at Broadcom Inc. and Nvidia Corp. have partnered to make AI more accessible, flexible and powerful than ever before, ensuring both variables expand in commensurate degrees. Nvidia and VCF are doing this by pushing the boundaries of computing through the seamless integration of hardware and software innovation.

Unwrapping next-gen AI infrastructure with Nvidia’s Manuvir Das and VMware’s Krish Prasad.

“A year ago, these models that are the basis of AI were just not good enough for enterprises to actually adopt in use cases,” said Manuvir Das (pictured, left), vice president of enterprise computing at Nvidia. “That’s fundamentally changed in the last year. These models have become really, really good. Now the challenge is, how are they going to do that? We’ve come to the point where enterprise IT needs to deploy a new platform.”

Das and Krish Prasad (right), senior vice president and general manager of VCF, spoke with theCUBE Research’s Dave Vellante and John Furrier at VMware Explore, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed the Nvidia and VCF partnership as a shared vision for the future of AI in the enterprise. (* Disclosure below.)

Nvidia and VCF: Virtualization meets AI with seamless integration

The virtualized GPU environment, initially announced as a work-in-progress, is now a key component in Nvidia and VMware’s shared portfolio, enabling enterprises to adopt AI with greater ease and efficiency. The joint solution from Nvidia and VCF is designed to offer a flexible platform that supports different AI models, whether on-premises or in the cloud, according to Das.

“I think customers want choice,” he said. “For some people, it’s okay to use somebody else’s service. For other customers, they need to have their private way of doing AI. The solution we’ve built allows a customer to say, ‘I’ll choose my environment.'”

At the core of this partnership is the virtualization concept. VMware’s vSphere with Tanzu and Nvidia’s AI Enterprise suite work together to virtualize AI workloads, allowing enterprises to run AI applications on existing infrastructure without major modifications. This approach reduces costs and simplifies the deployment process, according to Prasad.

“AI has now moved from the research labs to mainstream in the enterprise,” he said. “Over the past four years, we both have been really working with customers to make AI real in the enterprise. What we are seeing now is it’s actually happening with this use case. What customers want to do is not have a siloed separate environment to run AI.”

The integration of Nvidia’s GPU technology with VMware’s software stack enables enterprises to achieve AI performance on par with or better than bare metal servers, Prasad added. This capability is crucial for enterprises that want to maintain control over data while leveraging the power of AI.

“[Customers] want the same environment they use for other applications with all the controls and governance that they put on the other applications to apply to AI also,” he said. “That’s where VMware comes in, because we are the infrastructure on which they run all their applications. And now they can run AI on top of our infrastructure.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE Research’s coverage of VMware Explore: 

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

Photo: SiliconANGLE