NVIDIA at Hot Chips 35

The NVIDIA Hopper™ architecture is powering the next generation of accelerated computing with unprecedented performance, scalability, and security for every data center. With the ability to securely scale diverse workloads—from small enterprise to exascale HPC and trillion-parameter AI—Hopper enables brilliant innovators to fulfill their life's work at the fastest pace in human history.

The NVIDIA Grace™ CPU delivers high performance, power efficiency, and high-bandwidth connectivity that can be used for HPC and AI applications. NVIDIA Grace Hopper Superchip is a breakthrough integrated CPU+GPU for giant-scale AI and HPC applications. For CPU-only HPC applications, the NVIDIA Grace CPU Superchip provides the highest performance, memory bandwidth, and energy efficiency compared to today’s leading server chips.

The NVIDIA® BlueField® data processing unit (DPU) ignites unprecedented innovation for data centers and supercomputing infrastructures. By offloading, accelerating, and isolating a broad range of advanced networking, storage, and security services, BlueField DPUs provide a secure and accelerated infrastructure for any workload—in any environment, from the cloud to the data center to the edge.

Explore DGX H100 with Jensen Huang as your guide. Check out the accelerated computing DGX engine behind the large language model (LLM) breakthrough, and learn why the NVIDIA DGX platform is the blueprint for half of the Fortune 100 customers building AI Infrastructure worldwide. 

Increasing compute demands in AI and HPC—including an emerging class of trillion-parameter models—are driving a need for multi-node, multi-GPU systems with seamless, high-speed communication between every GPU. To build the DGX GH200 computing platform, a fast, scalable interconnect—the NVLink Switch System—is employed to offer 144 terabytes (TB) of shared memory with linear scalability for giant AI models.

The NVIDIA H100 Tensor Core GPU is our ninth-generation data center GPU designed to deliver an order-of-magnitude performance leap for large-scale AI and HPC over the prior-generation NVIDIA A100 Tensor Core GPU. Get an in-depth look at the new NVIDIA H100 GPU and the new features of the NVIDIA Hopper architecture.

The NVIDIA® GH200 Grace Hopper architecture brings together the groundbreaking performance of the NVIDIA Hopper GPU with the versatility of the NVIDIA Grace™ CPU, connected with a high bandwidth and memory coherent NVIDIA NVLink® Chip-2-Chip (C2C) interconnect in a single superchip, and support for the new NVLink Switch System. Learn more from this deep dive into the NVIDIA Grace Hopper Superchip Architecture.

The NVIDIA Spectrum™-X networking platform, featuring NVIDIA Spectrum-4 switches and BlueField®-3 data processing units (DPUs), is the world’s first ethernet fabric built for AI, accelerating generative AI performance by 1.7X over traditional ethernet fabrics. With Spectrum-X, cloud service providers can accelerate the development, deployment, and time to market of AI solutions, while improving return on investment.

Apply to try vSphere on BlueField with NVIDIA LaunchPad. You’ll get access to VMware ESXi-optimized software running on BlueField DPU-accelerated infrastructure for prototyping and testing next-generation applications and workloads.