NVIDIA has announced the NVIDIA DGX GH200, a groundbreaking AI supercomputer. Powered by the revolutionary NVIDIA GH200 Grace Hopper superchip and NVIDIA NVLink switch system.
This supercomputer is designed to facilitate the development of state-of-the-art generative AI language applications, recommender systems and data analytics workloads.
NVIDIA DGX GH200 boasts extensive shared memory space leveraging NVLink interconnect technology with NVLink switch systems. Combining 256 GH200 superchips, these supercomputers can effectively operate as a single GPU. This remarkable configuration delivers a whopping 1 exaflop of performance and a staggering 144 terabytes of shared memory, nearly 500 times the memory capacity of his previous generation NVIDIA DGX A100, which launched in 2020. To do.
NVIDIA founder and CEO Jensen Huang emphasized the importance of generative AI, large-scale language models, and recommender systems as powerhouses of the digital economy. “The DGX GH200 AI supercomputer integrates NVIDIA’s leading-edge accelerated computing and networking technologies to push the boundaries of AI,” he said. NVIDIA NVLink Technology Scales AI at Scale
The GH200 Superchip combines an Arm-based NVIDIA Grace CPU and an NVIDIA H100 Tensor Core GPU in the same package using the NVIDIA NVLink-C2C chip interconnect, obviating the need for traditional CPU-to-GPU PCIe connectivity. Eliminate. This delivers a 7x increase in GPU-to-CPU bandwidth and more than a 5x reduction in interconnect power consumption compared to the latest PCIe technology, delivering a 600 GB hopper architecture GPU building block for the DGX GH200 supercomputer will be
The DGX GH200 is the first supercomputer to combine the Grace Hopper superchip with the NVIDIA NVLink switch system. This is a new interconnect that allows all the GPUs in a DGX GH200 system to work together as one. In previous generation systems, he was able to combine 8 GPUs as 1 GPU with NVLink without compromising performance.
The DGX GH200 architecture delivers 48x more NVLink bandwidth than the previous generation, delivering the power of massive AI supercomputers with the simplicity of programming a single GPU.
New research tools for AI pioneers
Google Cloud, Meta, and Microsoft are expected to be the first to get access to the DGX GH200 to explore the capabilities of generative AI workloads. NVIDIA also plans to make the DGX GH200 design available as a blueprint to cloud service providers and other hyperscalers for further customization to their infrastructure.
“Building sophisticated generative models requires innovative approaches to AI infrastructure,” said Mark Lohmeyer, VP of Computing at Google Cloud. “Grace Hopper’s new NVLink scale and shared memory on superchip addresses a major bottleneck in large-scale AI, and we look forward to exploring its capabilities on Google Cloud and our generative AI initiatives.” I have.”
Alexis Björlin, Vice President of Infrastructure, AI Systems and Accelerated Platforms at Meta, said: “His Grace Hopper design at NVIDIA is intended to enable researchers to explore new approaches to solving their biggest challenges.”
Girish Bablani, Corporate Vice President of Azure Infrastructure at Microsoft, said: “The potential of the DGX GH200 to handle terabyte-sized datasets will enable developers to conduct advanced research at greater scale and accelerated speed.”
New NVIDIA Helios Supercomputer Drives R&D
NVIDIA is building its own DGX GH200-based AI supercomputer to power the work of researchers and development teams.
Dubbed NVIDIA Helios, the supercomputer will feature four DGX GH200 systems. Each is interconnected with his NVIDIA Quantum-2 InfiniBand networking to increase data throughput for training large-scale AI models. Helios contains 1,024 Grace Hopper Superchips and will be online by the end of the year.
Fully integrated and purpose-built for giant models
The DGX GH200 supercomputer includes NVIDIA software that provides a turnkey, full-stack solution for the largest AI and data analytics workloads. NVIDIA Base Command software provides AI workflow management, enterprise-grade cluster management, libraries to accelerate compute, storage, and network infrastructure, and system software optimized to run AI workloads.
It also includes NVIDIA AI Enterprise, the software layer of the NVIDIA AI platform. It offers over 100 frameworks, pre-trained models, and development tools to streamline the development and deployment of production AI, including generative AI, computer vision, speech AI, and more.
availability
The NVIDIA DGX GH200 supercomputer is expected to be available by the end of the year.
NVIDIA Announces Quake II RTX Version 1.7.0 With Enhanced Features – 04/19/2023 08:42 AM
NVIDIA has released Quake II RTX version 1.7.0 with a variety of exciting enhancements and important bug fixes. Quake II RTX is NVIDIA’s popular tech demonstration platform and…
NVIDIA Announces New Workstation GPU Series: RTX Ada Including RTX 4000 SFF ADA Desktop GPU and 5 Mobile SKUs – 2023/03/21 7:11 PM
The new desktop GPU RTX 4000 ADA SFF is a small form factor professional GPU boasting impressive specs of 6144 CUDA cores, 192 Tensors and 48 RT cores, delivering up to 19.2 TFLOPS of compute per second. increase.
NVIDIA Announces DGX Cloud, Revolutionary AI Supercomputing Service for the Enterprise – 2023/03/21 7:03 PM
NVIDIA launches DGX Cloud, a revolutionary AI supercomputing service. It provides unparalleled access to state-of-the-art infrastructure and software for companies to develop advanced generative systems.
NVIDIA Announces Game Bundle Featuring Redfall: New Open-World First-Person Shooter – 2023/03/15 7:07 PM
NVIDIA will be announcing a new game bundle containing a special edition of Redfall, the new game developed by Arkane Studios and published by Bethesda.
NVIDIA announces Grace CPU. It has 144 cores and 1 TB/s bandwidth – 2022/03/22 8:21 PM
If you watched today’s presentation, you may have also noticed the announcement of a CPU processor called “Grace.” …
Click here to post your comments on this news article on the message forum.
