AI factory management is expanding beyond software and servers to encompass the physical systems supporting production workloads. Broadcom Inc. and Super Micro Computer Inc. are integrating their technologies to coordinate artificial intelligence infrastructure from hardware provisioning through lifecycle operations.

VMware AI Factory supplies the software-defined layer, while Supermicro’s management suite extends visibility into servers, networking, power, cooling and firmware. The expanded partnership is intended to give enterprises and cloud providers a more unified way to deploy and operate large AI environments.

“I wanted to … tell you why Supermicro for AI, why VMware and why we are peanut butter and jelly together,” said Somik Behera (pictured, right), general manager of cloud, datacenter and AI software products at Supermicro. “Together, every enterprise gets a one-stop solution: a single unified integrated solution across storage, compute, AI and an emerging AI-native application development environment.”

Behera and Vijay Ramachandran (left), vice president of product management and core infrastructure at Broadcom, spoke with theCUBE Research’s Christophe Bertrand and co-host Alison Kosik at VMware Explore, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how integrated infrastructure can reduce deployment complexity as AI shifts toward enterprise applications. (* Disclosure below.)

AI factory management extends from workloads to cooling

VMware Cloud Foundation and Supermicro’s HGX systems provide complementary management layers. VMware automates software deployment and lifecycle operations, while SuperCloud Director, SuperCloud Automation Center and SuperCloud Composer manage physical infrastructure across multitenant environments, Behera noted.

“The approach that we have taken with the VMware AI Factory is a software-defined approach,” he said. “There’s no dependency on specific hardware. With that approach, we can expand this AI factory to any certified hardware vendor, with specific validation across various hardware vendors.”

That integration targets enterprises seeking graphics processing unit capacity through neocloud providers as training gives way to inference and application development, according to Behera. It also extends Broadcom’s broader effort to connect private AI infrastructure, software and governance.

“The neocloud started off with the AI labs doing training, but the training needs to result in money, which means you have to build applications, workflows, drive business outcomes,” Behera said. “And guess who does that? Enterprises. These enterprises are now getting held back because they do not have GPU capacity; they don’t have a turnkey solution to move their workloads to these next-generation GPUs. With our partnership, they can take this validated, preconfigured AI factory.”

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

(* Disclosure: TheCUBE is a paid media partner for the VMware Explore event. Sponsors of theCUBE’s event coverage do not have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE