"How do these 'snort your coffee' numbers arise?": Expert questions the validity of Zettascale and Exascale-class AI supercomputers, and presents a simple compelling car analogy to explain not all FLOPs are the same

supercomputer
(Image credit: Shutterstock / Timofeev Vladimir)

A leading expert has raised critical questions about the validity of claims surrounding "Zettascale" and "Exascale-class" AI supercomputers.

In an article that delves deep into the technical intricacies of these terms, Doug Eadline from HPCWire explains how terms like exascale, which traditionally denote computers achieving one quintillion floating-point operations per second (FLOPS), are often misused or misrepresented, especially in the context of AI workloads.

Eadline points out that many of the recent announcements touting "exascale" or even "zettascale" performance are based on speculative metrics, rather than tested results. He writes, "How do these 'snort your coffee' numbers arise from unbuilt systems?" - a question that highlights the gap between theoretical peak performance and actual measured results in the field of high-performance computing. The term exascale has historically been reserved for systems that achieve at least 10^18 FLOPS in sustained, double-precision (64-bit) calculations, a standard verified by benchmarks such as the High-Performance LINPACK (HPLinpack).

Car comparison

As Eadline explains, the distinction between FLOPS in AI and HPC is crucial. While AI workloads often rely on lower-precision floating-point formats such as FP16, FP8, or even FP4, traditional HPC systems demand higher precision for accurate results.

The use of these lower-precision numbers is what leads to inflated claims of exaFLOP or even zettaFLOP performance. According to Eadline, "calling it 'AI zetaFLOPS' is silly because no AI was run on this unfinished machine."

He further emphasizes the importance of using verified benchmarks like HPLinpack, which has been the standard for measuring HPC performance since 1993, and how using theoretical peak numbers can be misleading.

The two supercomputers that are currently part of the exascale club - Frontier at Oak Ridge National Laboratory and Aurora at Argonne National Laboratory - have been tested with real applications, unlike many of the AI systems making exascale claims.

To explain the difference between various floating-point formats, Eadline offers a car analogy: "The average double precision FP64 car weighs about 4,000 pounds (1814 Kilos). It is great at navigating terrain, holds four people comfortably, and gets 30 MPG. Now, consider the FP4 car, which has been stripped down to 250 pounds (113 Kilos) and gets an astounding 480 MPG. Great news. You have the best gas mileage ever! Except, you don’t mention a few features of your fantastic FP4 car. First, the car has been stripped down of everything except a small engine and maybe a seat. What’s more, the wheels are 16-sided (2^4) and provide a bumpy ride as compared to the smooth FP64 sedan ride with wheels that have somewhere around 2^64 sides. There may be places where your FP4 car works just fine, like cruising down Inference Lane, but it will not do well heading down the FP64 HPC highway."

Eadline’s article serves as a reminder that while AI and HPC are converging, the standards for measuring performance in these fields remain distinct. As he puts it, "Fuzzing things up with 'AI FLOPS' will not help either," pointing out that only verified systems that meet the stringent requirements for double-precision calculations should be considered true exascale or zettascale systems.

More from TechRadar Pro

Wayne Williams
Editor

Wayne Williams is a freelancer writing news for TechRadar Pro. He has been writing about computers, technology, and the web for 30 years. In that time he wrote for most of the UK’s PC magazines, and launched, edited and published a number of them too.

Read more
Project DIGITS - front view
I am thrilled by Nvidia’s cute petaflop mini PC wonder, and it’s time for Jensen’s law: it takes 100 months to get equal AI performance for 1/25th of the cost
A hand reaching out to touch a futuristic rendering of an AI processor.
DeepSeek and the race to surpass human intelligence
A person holding out their hand with a digital AI symbol.
AI smartphone and laptop sales are said to be slowly dying – but is anyone surprised?
Cerebras WSE-3
DeepSeek on steroids: Cerebras embraces controversial Chinese ChatGPT rival and promises 57x faster inference speeds
Nvidia H800 GPU
A look at the unbelievable Nvidia GPU that powers DeepSeek's AI global ambition
The Ryzen AI Max+ 395 could power the latest generation of powerful mini PCs
The AMD Ryzen AI Max+ 395 dominates as the "most powerful" APU on the market, but its competition is questionable
Latest in Pro
Epson EcoTank ET-4850 next to a TechRadar badge that reads Big Savings
I found the best printer deal you won't see in the Amazon Spring Sale and it's got a massive $150 saving
NVIDIA RTX PRO 6000 Blackwell Server Edition
Nvidia's most expensive Blackwell card gets massive price cut but it is not the RTX 5090
Microsoft Copiot Studio deep reasoning and agent flows
Microsoft reveals OpenAI-powered Copilot AI agents to bosot your work research and data analysis
Group of people meeting
Inflexible work policies are pushing tech workers to quit
Data leak
Top home hardware firm data leak could see millions of customers affected
Representational image depecting cybersecurity protection
Third-party security issues could be the biggest threat facing your business
Latest in News
Buzz Lightyear Space Ranger Spin Rennovations
Disney’s giving a classic Buzz Lightyear ride a tech overhaul – here's everything you need to know
Hisense U8 series TV on wall in living room
Hisense announces 2025 mini-LED TV lineup, with screen sizes up to 100 inches – and a surprising smart TV switch
Nintendo Music teaser art
Nintendo Music expands its library with songs from Kirby and the Forgotten Land and Tetris
Opera AI Tabs
Opera's new AI feature brings order to your browser tab chaos
An image of Pro-Ject's Flatten it closed and opened
Pro-Ject’s new vinyl flattener will fix any warped LPs you inadvertently buy on Record Store Day
The iPhone 16 Pro on a grey background
iPhone 17 Pro tipped to get 8K video recording – but I want these 3 video features instead