NVIDIA Introduces HGX-2, Fusing HPC and AI Computing
NVIDIA today introduced NVIDIA HGX-2, the first unified computing platform for both artificial intelligence and high performance computing.
The HGX-2 cloud server platform, with multi-precision computing capabilities, provides unique flexibility to support the future of computing. It allows high-precision calculations using FP64 and FP32 for scientific computing and simulations, while also enabling FP16 and Int8 for AI training and inference. This unprecedented versatility meets the requirements of the growing number of applications that combine HPC with AI.
A number of leading computer makers today shared plans to bring to market systems based on the NVIDIA HGX-2 platform.
“The world of computing has changed,” said Jensen Huang, founder and chief executive officer of NVIDIA, speaking at the GPU Technology Conference Taiwan, which kicked off today. “CPU scaling has slowed at a time when computing demand is skyrocketing. NVIDIA’s HGX-2 with Tensor Core GPUs gives the industry a powerful, versatile computing platform that fuses HPC and AI to solve the world’s grand challenges.”
HGX-2-serves as a “building block” for manufacturers to create some of the most advanced systems for HPC and AI. It has achieved record AI training speeds of 15,500 images per second on the ResNet-50 training benchmark, and can replace up to 300 CPU-only servers.
HGX-1 | HGX-2 | |
---|---|---|
Performance | 1 petaFLOP tensor operations 125 teraFLOPS single-precision 62 teraFLOPS double-precision |
2 petaFLOPS tensor operations 250 teraFLOPS single-precision 125 teraFLOPS double-precision |
GPUs | 8x NVIDIA Tesla V100 | 16x NVIDIA Tesla V100 |
GPU Memory | 256GB total | 512GB total |
NVIDIA CUDA® Cores | 40,960 | 81,920 |
NVIDIA Tensor Cores | 5,120 | 10,240 |
Communication Channel | Hybrid cube mesh powered by NVLink 300GB/s bisection bandwidth | NVSwitch powered by NVLink 2.4TB/s bisection bandwidth |
It incorporates such breakthrough features as NVIDIA NVSwitch™ interconnect fabric, which seamlessly links 16 NVIDIA Tesla® V100 Tensor Core GPUs to work as a single, giant GPU delivering two petaflops of AI performance. The first system built using HGX-2 was the recently announced NVIDIA DGX-2™.
HGX-2 comes a year after the launch of the original NVIDIA HGX-1, at Computex 2017. The HGX-1 reference architecture won broad adoption among the world’s leading server makers and companies operating massive datacenters, including Amazon Web Services, Facebook and Microsoft.
OEM, ODM Systems Expected Later This Year
Four leading server makers — Lenovo, QCT, Supermicro and Wiwynn — announced plans to bring their own HGX-2-based systems to market later this year.
Additionally, four of the world’s top original design manufacturers (ODMs) — Foxconn, Inventec, Quanta and Wistron — are designing HGX-2-based systems, also expected later this year, for use in some of the world’s largest cloud datacenters.
Family of NVIDIA GPU-Accelerated Server Platforms
HGX-2 is a part of the larger family of NVIDIA GPU-Accelerated Server Platforms, an ecosystem of qualified server classes addressing a broad array of AI, HPC and accelerated computing workloads with optimal performance.
Supported by major server manufacturers, the platforms align with the datacenter server ecosystem by offering the optimal mix of GPUs, CPUs and interconnects for diverse training (HGX-T2), inference (HGX-I2) and supercomputing (SCX) applications. Customers can choose a specific server platform to match their accelerated computing workload mix and achieve best-in-class performance.
Nvidia is Teasing Something To Be Released 29.04.17 - Updated Myth Busted! - 04/12/2017 05:11 PM
On a regional Nvidia facebook page Nvidia is teasing that something is to be announced or released April the 29th, which would be on a Saturday....
NVIDIA introduces Xavier SoC based on Volta - for Autonomous Transportation - 09/28/2016 05:06 PM
Xavier is a complete system-on-chip (SoC), integrating a new GPU architecture called Volta, a custom 8 core CPU architecture, and a new computer vision accelerator. The processor will deliver 20 TOPS ...
AMD Takes Back Market Share from NVIDIA in Q3 2015 - 11/16/2015 10:01 AM
Jon Peddie Research (JPR), the industry's research and consulting firm for graphics and multimedia, announced estimated PC graphics add-in-board (AIB) shipments and suppliers' market share for Q3'1...
Nvidia Intros NVS 810 Business videocard with eight DisplayPort connectors - 11/05/2015 11:22 AM
Nvidia has released the NVS 810 Business videocard, this puppy holds no less than eight DisplayPort connectors each capable of driving an Ultra HD screen. This single slot card is powered by two GM107...
NVIDIA Interview: The Sky Isn't Falling - 09/24/2013 06:07 PM
I'm not sure why people feel the need to explain that it's "no longer possible" for consoles to have better graphics than a PC". Have a peek at a part of the interview, after the b...
Member
Posts: 66
Joined: 2006-09-23
am I missing something, isn't HGX-2 just 2x HGX-1s? The only new item is the NVLink Communication Channel.
Senior Member
Posts: 1502
Joined: 2013-10-31
Okay this is great but when can we get next gen cards?
Senior Member
Posts: 214
Joined: 2014-06-11
The power to create the Death Star is nigh upon us!
Senior Member
Posts: 14623
Joined: 2014-07-21
I think this could run Crysis even at 60fps
Very much performance. Powerhouse. I don't want to know how to cool this thing.