BlueField-2: Nvidia Debuts Data Center Services Accelerator

In his keynote address at the GPU Technology Conference (GTC), Nvidia CEO Jensen Huang stated that the company shipped more than one billion GPUs and its CUDA software development kit had six million downloads thus far in 2020. He revealed 80 new SDKs and 1,800 GPU-accelerated applications are available. The company also unveiled its BlueField-2, a powerful data center services accelerator. GTC was expected to draw 30,000 global attendees this week for more than 1,000 sessions and coincides with the Arm DevSummit, which features a chat between Huang and Arm chief executive Simon Segars.

VentureBeat reports that head of enterprise computing Manuvir Das introduced the Nvidia BlueField-2 — a data processing unit (DPU) — aimed at making it easier to run cloud-based data centers. DPUs “combine Nvidia’s chip technology with the networking, security, and storage technology the company gained with its $7.5 billion acquisition of Mellanox in 2019.”

The BlueField-2 DPU (above) is comprised of “eight 64-bit A72 ARM cores.” According to Das, BlueField-2 “provides accelerated data center infrastructure services in which central processing units (CPUs), GPUs, and DPUs work together to deliver a computing unit that is AI-enabled, programmable, and secure.”

“We believe that the DPU belongs in every server going forward, regardless of the application workload running there,” he said.

According to Nvidia, “a single BlueField-2 DPU can deliver the same data center services that might consume up to 125 CPU cores … [which] frees up valuable CPU cores to run a wide range of other enterprise applications.” It can “handle 0.7 trillion operations per second (TOPS), while the BlueField-2X with its Ampere GPU can do 60 TOPS.”

With DPUs, says Nvidia, “future enterprise data centers will have one or more servers across 10,000 different locations, including inside office buildings, factories, warehouses, cell towers, schools, stores, and banks.” These “edge data centers” can support the Internet of Things. To “simplify and secure the deployment and management of AI applications and models on these servers at scale,” Nvidia announced early access program to Nvidia Fleet Command. Kion Group, a supply chain company, and Northwestern Memorial Hospital in Illinois are two of the first companies to use Fleet Command.

Nvidia also debuted the latest version of Nvidia Jetson AI at the Edge, to be available end of October priced at $59 and up and aimed at “students, educators, and robotics hobbyists.” The Jetson Nano 2GB Developer Kit, also priced at $59, “is designed for teaching and learning AI through hands-on projects in such areas as robotics and intelligent Internet of Things.”

Nvidia also revealed that, “its Ampere-based Nvidia RTX A6000 workstation chips will replace the Turing-based version of the Quadro family,” and a new Nvidia A40 is the “passive-cooled version of the same chip.” Professional virtual reality director David Weinstein stated Nvidia “will enable CloudXR on Amazon Web Services, where virtual reality (VR), augmented reality (AR), and mixed reality (XR) experiences can be streamed to VR headsets via the cloud.”

To watch Huang describe the Nvidia BlueField-2 DPU, click here.