Nvidia HGX Vs DGX: What Are The Differences?

How Nvidia’s enterprise GPU solutions for high-performance AI computing differ Source: Lenovo Key Takeaways Nvidia currently dominates the AI hardware market with DGX and HGX platforms tailored for different enterprise … Read more

Taylor Bell

Taylor Bell

Published on Apr 07, 2024

Nvidia HGX Vs DGX: What Are The Differences?

How Nvidia’s enterprise GPU solutions for high-performance AI computing differ

Multiple datacenter racks with Lenovo ThinkStation PX workstations

Source: Lenovo

Key Takeaways

  • Nvidia currently dominates the AI hardware market with DGX and HGX platforms tailored for different enterprise needs.
  • Nvidia DGX comprises systems and cluster solutions based around Nvidia’s Hopper and Blackwell GPU architectures.
  • Nvidia HGX offers the same underlying hardware, but in a modular and customizable package.

Nvidia is comfortably riding the AI wave. And for at least the next few years, it will likely not be dethroned as the AI hardware market leader. With its extremely popular enterprise solutions powered by the H100 and H200 “Hopper” lineup of GPUs (and now B100 and B200 “Blackwell” GPUs), Nvidia is the go-to manufacturer of high-performance computing (HPC) hardware.

Nvidia DGX is an integrated AI HPC solution targeted toward enterprise customers needing immensely powerful workstation and server solutions for deep learning, generative AI, and data analytics. Nvidia HGX is based on the same underlying GPU technology. However, HGX is a customizable enterprise solution for businesses that want more control and flexibility over their AI HPC systems. But how do these two platforms differ from each other?

Nvidia“” data-modal-id=”single-image-modal” data-modal-container-id=”single-image-modal-container” data-img-caption=””””>

An image showing a render of the Nvidia DGX SuperPod system.

Related

What is Nvidia DGX platform?

Nvidia’s standardized HPC platform for enterprise AI applications.

Nvidia DGX: The original supercomputing platform

The “green” standard

It should surprise no one that Nvidia’s primary focus isn’t on its GeForce lineup of gaming GPUs anymore. Sure, the company enjoys the lion’s share among the best gaming GPUsbut its recent resounding success is driven by enterprise and data center offerings and AI-focused workstation GPUs.

The Nvidia DGX platform integrates up to 8 Tensor Core GPUs with Nvidia’s AI software to power accelerated computing and next-gen AI applications. It’s essentially a rack-mount chassis containing 4 or 8 GPUs connected via NVLink, high-end x86 CPUs, and a bunch of Nvidia’s high-speed networking hardware. A single DGX B200 system is capable of 72 petaFLOPS of training and 144 petaFLOPS of inference performance.

The company currently offers both Hopper-based (DGX H100) and Blackwell-based (DGX B200) systems optimized for AI workloads. Customers can go a step further with solutions like the DGX SuperPOD (with DGX GB200 systems) that integrates 36 liquid-cooled Nvidia GB200 Grace Blackwell Superchips, comprised of 36 Nvidia Grace CPUs and 72 Blackwell GPUs. This monstrous setup includes multiple racks connected through Nvidia Quantum InfiniBand, allowing companies to scale thousands of GB200 Superchips.

Nvidia has been selling DGX systems for quite some time now — from the DGX Server-1 dating back to 2016 to modern DGX B200-based systems. From the Pascal and Volta generations to the Ampere, Hopper, and Blackwell generations, Nvidia’s enterprise HPC business has pioneered numerous innovations and helped in the birth of its customizable platform, Nvidia HGX.

Nvidia HGX: For businesses that need more

Build your own supercomputer

Nvidia“” data-modal-id=”single-image-modal” data-modal-container-id=”single-image-modal-container” data-img-caption=”null”>

HGX B200 tensor core GPU

For OEMs looking for custom supercomputing solutions, Nvidia HGX offers the same peak performance as its Hopper and Blackwell-based DGX systems but allows OEMs to tweak it as needed. For instance, customers can modify the CPUs, RAM, storage, and networking configuration as they please. Nvidia HGX is actually the baseboard used in the Nvidia DGX system, but adheres to Nvidia’s own standard.

Nvidia offers Nvidia HGX in x4 and x8 GPU configurations, with the latest Blackwell-based baseboards only available in the x8 configuration. With up to 144 petaFLOPS of performance, an HGX B200 system can supercharge enterprise AI and HPC computing, while allowing OEMs to build scalable solutions on it.

Nvidia DGX vs. HGX: Summary

Simplicity vs. Flexibility

Nvidia's H100 Tensore Core GPU-based NVL shown on a blue background

Source: Nvidia

  • While Nvidia DGX represents Nvidia’s line of standardized, unified, and integrated supercomputing solutions, Nvidia HGX unlocks greater customization and flexibility for OEMs to offer more to enterprise customers.
  • With Nvidia DGX, the company leans more into cluster solutions that integrate multiple DGX systems into huge and, in the case of the DGX SuperPOD, multi-million-dollar data center solutions. Nvidia HGX, on the other hand, is another way of selling HPC hardware to OEMs at a greater profit margin.
  • Nvidia DGX brings rapid deployment and a seamless, hassle-free setup for bigger enterprises. Nvidia HGX provides modular solutions and greater access to the wider industry.

The next step in the AI wars

Nvidia looks to dominate the AI hardware market in the near future. But, it might be facing a challenge as Google, Intel, and Qualcomm plan to reduce the industry’s reliance on Nvidia hardware. These companies, part of the UXL Foundation, are uniting to build a suite of software and tools under “OneAPI” to allow AI code to run on various kinds of AI hardware, not just that from Nvidia. Meanwhile, Team Green is constantly adapting to find newer software applications for its hardware, such as Chat with RTX.

For the time being, however, Nvidia’s iron grip on the hardware side of the AI revolution will likely stay unchanged.

Partager cet article

Inscrivez-vous à notre newsletter