Nvidia, Amazon Tease 6x Performance Boost To Upcoming Supercomputer

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

Find products, advice, tech news

PCMag Middle East Computers & Electronics Components

Nvidia, Amazon Tease 6x


Performance Boost to
Upcoming Supercomputer
The companies revise the Project Ceiba
supercomputer to have six times more computing
power than originally envisioned thanks to Nvidia's
new Blackwell GPU architecture.

by Michael Kan
Mar 19, 2024

(CFOTO/Future Publishing via Getty Images)

Nvidia’s ambitions to build one of the world’s


fastest supercomputers are becoming even
grander.

The upcoming supercomputer, dubbed Project


Ceiba, was originally announced in November in
partnership with Amazon. But on Monday, the
companies announced they planned on
upgrading the machine with a “6x performance
increase.”

Nvidia will harness its newly announced


Blackwell GPU architecture, the successor to its
Hopper-based H100 GPUs. Swapping in
Blackwell will enable the supercomputer to reach
a processing power of 414 exaflops, up from a
mere 65 exaflops.

To pull this off, Nvidia plans on installing 20,736


Blackwell B200 GPUs inside the supercomputer,
alongside 10,368 Grace CPUs. All that computing
power will be hosted through Amazon’s AWS
cloud service.

It remains unclear when Project Ceiba will be


completed. But we’re curious to see where the
supercomputer will rank once it’s up and
running.

The current most powerful supercomputer,


Frontier, can reach over 1.1 exaflops when
measured using the Linpack benchmark.
Microsoft’s cloud supercomputer Eagle, which
was built with 14,400 Nvidia H100 GPUs, ranks
third and is able to reach 0.56 exaflops.

Nvidia didn’t say how it arrived at the 414 exaflop


number. But the company probably used an AI
inference performance measurement rather
than a traditional supercomputing benchmark.

As for Project Ceiba, Nvidia plans on using the


machine for research and development,
including “advancing the application of AI to
graphics and simulation, digital biology, robotics,
self-driving cars, climate prediction, and more,”
the company said. The machine will also be used
to develop large language models, such as
OpenAI’s GPT-4.

MORE INSIDE PCMAG.COM

What Is a TPM, and Why Do I Need One for

Windows 11?

GlobalFoundries Gets $1.5 Billion From US to

Increase Domestic Chip Manufacturing

'Meteor Lake' Launches! Speeds & Feeds for

Intel Core Ultra Laptop CPUs Are Here

Silent Running: How to Fix a Noisy Computer

Fan

How Much RAM Do I Have in My PC?

About Michael Kan

I've been with PCMag since October

2017, covering a wide range of topics,

including consumer electronics, cybersecurity,

social media, networking, and gaming. Prior to

working at PCMag, I was a foreign

correspondent in Beijing for over five years,

covering the tech scene in Asia.

More From Michael Kan

Chase Bank to Let Advertisers Target

Customers Based on Spending Habits

Refreshed Razer Blade 18 Gaming Laptop

With Thunderbolt 5 Now on Sale

Badge of Shame? Twitter Restores Blue

Checkmarks: Here's How to Remove Them

WhatsApp Outage Prevents Users From

Sending, Receiving Messages

SpaceX Now Offering 'Expert Installs' of

Starlink for $199

COM MENTS

News About PC Mag Middle East

Features Privacy Policy

Gaming Terms of Use

Esports Cookies

Business

Motoring

How-To

Registration

‫يﺔ‬#‫اﻟﻌرٮ‬

PCMag is obsessed with culture and tech,


offering smart, spirited coverage of the products
and innovations that shape our connected lives
and the digital trends that keep us talking.

PCMag Middle East is operated under license by Sheeria


Group.

International Editions:

PCMag USA Greece

United Kingdom Israel

Australia Middle East

You might also like