AI workloads powered by Dell and AMD – News

The ongoing partnership between Advanced Micro Devices Inc. and Dell Technologies Inc. has manifested itself in a number of ways over the years. The most recent iteration involves Dell’s PowerEdge systems, which have been built using AMD’s fifth-generation Turin processor technology to support AI workloads.

The announcement of the Turin-equipped PowerEdge offering in November highlighted Dell’s interest in delivering artificial intelligence for compute that would provide purpose-built, large-scale AI solutions.

“When we sat down and first started discussing Turin with AMD, we recognized that there’s different sizes for different parts of customer journeys, what they need to deploy in their data center,” said David Schmidt (pictured, right), senior director, PowerEdge product management, at Dell. “We looked at how AMD was laying out Turin, the Turin stack from top to bottom, and we really challenged each other on what we needed to provide. We were bringing system design ideas to the table of serving customers that need eight cores, 16 cores. It’s not all about the highest core counts possible; it’s about providing the right range.”

Schmidt spoke with theCUBE’s Dave Vellante at the “Is Your IT Infrastructure Ready for the Age of AI?” event, during an exclusive broadcast on theCUBE, News Media’s livestreaming studio. He was joined by Derek Dicker (left), corporate vice president, enterprise and HPC business group, at AMD. They discussed how a partnership between the two firms has helped advance compute capabilities for customers. (* Disclosure below.)

Providing balance for AI workloads

Schmidt’s point about finding the right range speaks to the balance that systems providers such as Dell and processor suppliers such as AMD must find in building a solution that meets enterprise needs in handling AI workloads. Dell’s generation of servers based on AMD Turin includes PCIe Multi-Segment, a feature that provides scalability as part of the release, according to Dicker.

“It’s essentially 128 lanes of PCIe with the ability to bifurcate those down,” Dicker explained. “What we’ve been able to do with input from Dell is architect a system that allows us to deliver those 128 lanes. You can shove 64 in the front, 64 out the back [and] have a balanced network connection, but also have the ability to service your storage in the system.”

Dell’s PowerEdge servers using AMD’s Turin processors support 12-channel DDR5 memory. This feature helps provide fast, efficient data handling to support data-heavy applications increasingly found in AI workloads today, according to Schmidt.

“You’ve got memory to core ratios that have to keep scaling,” Schmidt said. “Workloads are scaling with the core counts, but it’s only because we’re able to design the systems that are providing the right amount of memory.”

Here’s the complete video interview, part of News’s and theCUBE’s coverage of the “Is Your IT Infrastructure Ready for the Age of AI?” event:

(* Disclosure: TheCUBE is a paid media partner for the “Is Your IT Infrastructure Ready for the Age of AI?” event. Neither Dell Technologies Inc., Advanced Micro Devices Inc., and Intel Corp., the sponsors of this part of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or News.)

Photo: News