NVIDIA Hopper in Full Manufacturing

World’s Main Pc Makers Dell Applied sciences, HPE, Lenovo, Supermicro, Plus Cloud Service Suppliers AWS, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure Constructing H100-Primarily based Choices; Availability Begins Subsequent Month

GTC—NVIDIA immediately introduced that the NVIDIA H100 Tensor Core GPU is in full manufacturing, with international tech companions planning in October to roll out the primary wave of services and products based mostly on the groundbreaking NVIDIA Hopper™ structure.

Unveiled in April, H100 is constructed with 80 billion transistors and advantages from a variety of know-how breakthroughs. Amongst them are the highly effective new Transformer Engine and an NVIDIA NVLink® interconnect to speed up the biggest AI fashions, like superior recommender methods and enormous language fashions, and to drive improvements in such fields as conversational AI and drug discovery.

“Hopper is the brand new engine of AI factories, processing and refining mountains of information to coach fashions with trillions of parameters which are used to drive advances in language-based AI, robotics, healthcare and life sciences,” mentioned Jensen Huang, founder and CEO of NVIDIA. “Hopper’s Transformer Engine boosts efficiency as much as an order of magnitude, placing large-scale AI and HPC inside attain of firms and researchers.”

Along with Hopper’s structure and Transformer Engine, a number of different key improvements energy the H100 GPU to ship the subsequent large leap in NVIDIA’s accelerated compute knowledge heart platform, together with second-generation Multi-Occasion GPU, confidential computing, fourth-generation NVIDIA NVLink and DPX Directions.

A five-year license for the NVIDIA AI Enterprise software program suite is now included with H100 for mainstream servers. This optimizes the event and deployment of AI workflows and ensures organizations have entry to the AI frameworks and instruments wanted to construct AI chatbots, advice engines, imaginative and prescient AI and extra.

World Rollout of Hopper

H100 allows firms to slash prices for deploying AI, delivering the identical AI efficiency with 3.5x extra power effectivity and 3x decrease complete price of possession, whereas utilizing 5x fewer server nodes over the earlier technology.

For patrons who wish to instantly attempt the brand new know-how, NVIDIA introduced that H100 on Dell PowerEdge servers is now obtainable on NVIDIA LaunchPad, which supplies free hands-on labs, giving firms entry to the newest {hardware} and NVIDIA AI software program.

Prospects can even start ordering NVIDIA DGX™ H100 methods, which embody eight H100 GPUs and ship 32 petaflops of efficiency at FP8 precision. NVIDIA Base Command™ and NVIDIA AI Enterprise software program energy each DGX system, enabling deployments from a single node to an NVIDIA DGX SuperPOD™ supporting superior AI improvement of enormous language fashions and different large workloads.

H100-powered methods from the world’s main pc makers are anticipated to ship within the coming weeks, with over 50 server fashions out there by the top of the yr and dozens extra within the first half of 2023. Companions constructing methods embody Atos, Cisco, Dell Applied sciences, Fujitsu, GIGABYTE, Hewlett Packard Enterprise, Lenovo and Supermicro.

Moreover, a number of the world’s main greater schooling and analysis establishments might be utilizing H100 to energy their next-generation supercomputers. Amongst them are the Barcelona Supercomputing Middle, Los Alamos Nationwide Lab, Swiss Nationwide Supercomputing Centre (CSCS), Texas Superior Computing Middle and the College of Tsukuba.

H100 Coming to the Cloud

Amazon Net Providers, Google Cloud, Microsoft Azure and Oracle Cloud Infrastructure might be among the many first to deploy H100-based cases within the cloud beginning subsequent yr.

“We sit up for enabling the subsequent technology of AI fashions on the newest H100 GPUs in Microsoft Azure,” mentioned Nidhi Chappell, normal supervisor of Azure AI Infrastructure. “With the developments in Hopper structure coupled with our investments in Azure AI supercomputing, we’ll have the ability to assist speed up the event of AI worldwide”

“By providing our clients the newest H100 GPUs from NVIDIA, we’re serving to them speed up their most advanced machine studying and HPC workloads,” mentioned Karan Batta, vp of product administration at Oracle Cloud Infrastructure. “Moreover, utilizing NVIDIA’s subsequent technology of H100 GPUs permits us to help our demanding inside workloads and helps our mutual clients with breakthroughs throughout healthcare, autonomous autos, robotics and IoT.”

NVIDIA Software program Assist

The superior Transformer Engine know-how of H100 allows enterprises to shortly develop giant language fashions with the next degree of accuracy. As these fashions proceed to develop in scale, so does the complexity, generally requiring months to coach.

To sort out this, a number of the world’s main giant language mannequin and deep studying frameworks are being optimized on H100, together with NVIDIA NeMo Megatron, Microsoft DeepSpeed, Google JAX, PyTorch, TensorFlow and XLA. These frameworks mixed with Hopper structure will considerably velocity up AI efficiency to assist prepare giant language fashions inside days or hours.

To study extra about NVIDIA Hopper and H100, watch Huang’s GTC keynote. Register for GTC without spending a dime to attend periods with NVIDIA and trade leaders.

Leave a Reply