• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

NVIDIA Introduces HGX-2, Fusing HPC and AI Computing into Unified Architecture


Editor & Senior Moderator
Staff member
Oct 9, 2007
40,154 (8.37/day)
Hyderabad, India
Processor AMD Ryzen 7 2700X
Motherboard ASUS ROG Strix B450-E Gaming
Cooling AMD Wraith Prism
Memory 2x 16GB Corsair Vengeance LPX DDR4-3000
Video Card(s) Palit GeForce RTX 2080 SUPER GameRock
Storage Western Digital Black NVMe 512GB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) Creative Sound Blaster Recon3D PCIe
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Microsoft Sidewinder X4
Software Windows 10 Pro
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.

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.

Broad Industry Support
Top OEMs and ODMs have voiced strong support for HGX-2:

"Foxconn has long been dedicated to hyperscale computing solutions and successfully won customer recognition. We're glad to work with NVIDIA for the HGX-2 project, which is the most promising solution to fulfill explosive demand from AI/DL."

- Ed Wu, corporate executive vice president at Foxconn and chairman at Ingrasys

"Inventec has a proven history of delivering high-performing and scalable servers with robust innovative designs for our customers who run some of the world's largest datacenters. By rapidly incorporating HGX-2 into our future designs, we'll infuse our portfolio with the most powerful AI solution available to companies worldwide."

- Evan Chien, head of IEC White Box Product Center, China Business Line Director, Inventec

"NVIDIA's HGX-2 ups the ante with a design capable of delivering two petaflops of performance for AI and HPC-intensive workloads. With the HGX-2 server building block, we'll be able to quickly develop new systems that can meet the growing needs of our customers who demand the highest performance at scale."

- Paul Ju, vice president and general manager of Lenovo DCG

"As a leading cloud enabler, Quanta is committed to developing solutions for the next generation of clouds for a variety of innovative use cases. As we have seen a multitude of AI applications on the rise, Quanta works closely with NVIDIA to ensure our clients benefit from the latest and greatest GPU technologies. We are thrilled to broaden our GPU compute portfolio with this critical enabler for AI clouds as an HGX-2 launch partner."

- Mike Yang, senior vice president, Quanta Computer, and president, QCT

"To help address the rapidly expanding size of AI models that sometimes require weeks to train, Supermicro is developing cloud servers based on the HGX-2 platform. The HGX-2 system will enable efficient training of complex models."

- Charles Liang, president and CEO of Supermicro

"We are very honored to work with NVIDIA as a partner. The demand for AI cloud computing is emerging in today's modern technology environment. I strongly believe the high performance and modularized flexibility of the HGX-2 system will make great contributions to various computing areas, ranging from academics and science to government applications."

- Jeff Lin, president of Enterprise Business Group, Wistron

"Wiwynn specializes in delivering hyperscale datacenter and cloud infrastructure solutions. Our collaboration with NVIDIA and the HGX-2 server building block will enable us to provide our customers with two petaflops of computing for computationally intensive AI and HPC workloads."

- Steven Lu, Vice President, Wiwynn

View at TechPowerUp Main Site
Aug 10, 2007
2,029 (0.42/day)
Austin TX
System Name Hexy
Processor Intel i9 7900x @5ghz (direct die liquid cooling)
Motherboard EVGA x299 Dark
Cooling Custom Liquid W/ 3x120 Radiator
Memory 32Gb Gskill (B-Die) @3600 CL16
Video Card(s) GTX 1080Ti Liquid Cooled
Storage 512GB Sabrent Rocket
Display(s) PX329 + PX247
Case Custom Lian Li V100L
Audio Device(s) Sennheiser HD58X + Sennheiser GSX 1000
Power Supply Corsair RM 750x
Mouse Zelotes C-18
Keyboard Bloody 845R
Software Windows 10 Pro
Benchmark Scores over there --->
As expected this was the only thing discussed in the key note.. Just remember, the more you buy.. The more you save! :rolleyes:

Classic Nvidia

Oh and the pricing you ask??
Dec 22, 2011
3,338 (1.02/day)
When virtual spaceships for an unfinished game can set you back $27000 these days, that isn't too bad.
Aug 13, 2010
4,749 (1.26/day)
As expected this was the only thing discussed in the key note.. Just remember, the more you buy.. The more you save! :rolleyes:

Classic Nvidia

Oh and the pricing you ask??
Yeah, AMD's deep learning servers are much cheaper.
Sep 15, 2016
405 (0.26/day)
I think you guys are missing the point... I have absolutely zero use for this machine but if I was a billionaire I would buy one just to have a massive nerdboner looking inside the chassis.
Dec 22, 2011
3,338 (1.02/day)
Oh for sure, this thing is basically the ultimate GPU, it's a 81,920 CUDA Core monster with 512GB of HBM2 VRAM, with bonkers levels of performance across whatever work load you want to throw at it.

Stick a glass window on the side, add some RGB lighting, and you win the i own Skynet competition.
Sep 15, 2007
3,847 (0.80/day)
Police/Nanny State of America
System Name More hardware than I use :|
Processor 4.7 8350 - 4.2 4560K - 4.4 4690K
Motherboard Sabertooth R2.0 - Gigabyte Z87X-UD4H-CF - AsRock Z97M KIller
Cooling Mugen 2 rev B push/pull - Hyper 212+ push/pull - Hyper 212+
Memory 16GB Gskill - 8GB Gskill - 16GB Ballistix 1.35v
Video Card(s) Xfire OCed 7950s - Powercolor 290x - Oced Zotac 980Ti AMP! (also have two 7870s)
Storage Crucial 250GB SSD, Kingston 3K 120GB, Sammy 1TB, various WDs, 13TB (actual capactity) NAS with WDs
Display(s) X-star 27" 1440 - Auria 27" 1440 - BenQ 24" 1080 - Acer 23" 1080
Case Lian Li open bench - Fractal Design ARC - Thermaltake Cube (still have HAF 932 and more ARCs)
Audio Device(s) Titanium HD - Onkyo HT-RC360 Receiver - BIC America custom 5.1 set up (and extra Klipsch sub)
Power Supply Corsair 850W V2 - EVGA 1000 G2 - Seasonic 500 and 600W units (dead 750W needs RMA lol)
Mouse Logitech G5 - Sentey Revolution Pro - Sentey Lumenata Pro - multiple wireless logitechs
Keyboard Logitech G11s - Thermaltake Challenger
Software I wish I could kill myself instead of using windows (OSX can suck it too).
Oh for sure, this thing is basically the ultimate GPU, it's a 81,920 CUDA Core monster with 512GB of HBM2 VRAM, with bonkers levels of performance across whatever work load you want to throw at it.

Stick a glass window on the side, add some RGB lighting, and you win the i own Skynet competition.
Irresponsible levels of perf?

All you'd need is a leather jacket to go with it and you can be the biggest douche bag on a stage. (Oops, someone at Nvidia already has that covered)
Nov 15, 2016
371 (0.25/day)
System Name Sillicon Nightmares
Processor Intel i7 9700KF 5ghz (no avx offset), 4.7ghz ring 1.374 vcore
Motherboard Asus Z390 Strix F
Cooling DEEPCOOL Gamer Storm CAPTAIN 360
Memory 2x8GB G.Skill Trident Z RGB 3733 16,18,18,36 2T
Video Card(s) ASUS GTX 1060 Strix 6GB OC, Core: 2202/2215, Vcore: 1.075v, Mem: 4909mhz (Sillicon Lottery Jackpot)
Storage Samsung 840 EVO 1TB SSD, WD Blue 1TB, Seagate 3TB, DoA PoS Nvme that makes me cry
Display(s) BenQ XL2430 1080p 144HZ + (2) Samsung SyncMaster 913v 1280x1024 75HZ + A Shitty TV For Movies
Case Deepcool Genome ROG Edition
Audio Device(s) Bunta Sniff Speakers From The Tip Edition
Power Supply Corsair AX860i/Cable Mod Cables
Mouse Logitech G602
Keyboard Shitty Dell Office Keyboard
Software Windows 10 x64
Benchmark Scores ~`13500 Firestrike (1st for my hardware, *De Ja Fool can s\/ck me softly)
Yeah, AMD's deep learning servers are much cheaper.
and much shittier, hence you can charge ridiculous prices for the top of the line product