NVIDIA, powerful A100 GPUs join the HGX server platform

NVIDIA, powerful A100 GPUs join the HGX server platform


NVIDIA has announced that the powerful A100 GPUs will be part of its HGX AI super-computing platform, which will be able to count on other new technologies. including 80GB memory modules, Infiniband 400G network and Magnum IO GPUDirect Storage software.

A100, which recently showed its potential given by the Ampere architecture surpassing the Titan V as the most powerful GPU in the OctaneBench benchmark, is available in two variants: one with 40GB of HBM2E and the other with 80GB. The most expensive model boasts the widest bandwidth in the world at over 2TB / s. Made with a 7nm process, the GPU has 54.2 billion transistors arranged in 6,912 shader units, 432 texture mapping units, 160 ROPs and 432 tensor cores.

Connecting these GPUs within the huge racks supercomputers are technologies like Magnum IO GPUDirect Storage, which reduces latency by allowing direct access between GPU memory and storage. Infiniband network up to 400Gb / s enables bandwidth of up to 1.64Pb / s per 2,048 port switch, with the ability to connect more than one million nodes.

Jensen Huang, founder and CEO of NVIDIA, said:

The HPC revolution began in academia and is rapidly spreading to a wide range of industries. Key dynamics are driving super exponential advances that have made HPC a useful tool for industries. NVIDIA's HGX platform offers researchers unprecedented high-performance computing to tackle the toughest problems facing industries.

Not exactly designed for home use, the HGX platform is used by companies such as General Electric, which simulates computational fluid dynamics to design new large gas turbines and jet engines. It will also be used to build the next generation supercomputer at the University of Edinburgh, optimized for computational particle physics, to analyze data from particle experiments, such as the Large Hadron Collider.

If you are looking for a good motherboard for your new configuration enough on AMD Ryzen processors, on Amazon you can find the Gigabyte X570 AORUS ELITE at an attractive price.

Could This Be a Big Risk for NVIDIA Stock?

In today's video I look at recent news for NVIDIA (NASDAQ:NVDA). The stock price is reaching new highs, but could this be increasing the risk for stockholders? Below I share a few highlights from the video. 

  • It was recently reported that Broadcom (NASDAQ:AVGO), Marvell (NASDAQ:MRVL), and MediaTek (OTC:MDTKF) are backing the NVIDIA acquisition of Arm. Each company has its reasons for supporting the acquisition. Still, the widespread consensus is that this acquisition will help accelerate innovation and adoption of the Arm-based technology, which can fuel the growth of numerous companies.
  • The excitement of the upcoming stock split and the backing for the Arm acquisition by major semiconductor companies has fueled the stock price to reach new levels. Now NVIDIA could risk seeing a downturn in stock if any negative news comes out toward the acquisition.
  • Long-term investors should remember stock prices go up and down. Trying to time the market can lead to missed opportunities.
  • Click the video below for my full thoughts and analysis. 

    *Stock prices used were the closed prices of June 28, 2021. The video was published on June 28, 2021. 

    This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.