Skip to main content
  1. Home
  2. Computing
  3. Legacy Archives

IBM’s Roadrunner supercomputer hits early retirement due to power usage

Add as a preferred source on Google
IBM Roadrunner Supercomputer
Image used with permission by copyright holder

The Los Alamos National Laboratory has pulled the plug on the IBM-built supercomputer Roadrunner after just five years of service as it has become too much of a power hog. 

The Roadrunner was the first supercomputer in the world to be able to handle one million billion calculations per second (petaflop). Back in June 2008, it was the fastest supercomputer. Five years later, it still held the number 22 spot on March 31, 2013, the day it was taken offline.

Recommended Videos

To achieve petaflop computing speeds, IBM combined two kinds of processors into one machine – a strategy that was unheard of at the time. The Roadrunner was powered by 6,563 dual-core processors from AMD, each of which were linked to a special graphics processor (PowerXCell 8i) that was originally designed for the Sony PlayStation 3, but tweaked for scientific computing.

The Roadrunner’s hybrid architecture was revolutionary. “[It] got everyone thinking in new ways about how to build and use a supercomputer,” said Gary Grider from the laboratory’s High Performance Computing Division. “Specialized processors are being included in new ways on new systems, and being used in novel ways. Our demonstration with Roadrunner caused everyone to pay attention.”

Unfortunately, the Roadrunner has become too power inefficient for the lab to keep in commission, especially as newer supercomputers with more computing prowess – and that also require less energy to power – become available. According to Ars Technica, the Roadrunner required 2,345 kilowatts to hit 1.042 petaflops. To give you a sense of how power hungry the Roadrunner has become next to its contemporaries, the supercomputer occupying the twenty-first spot needs just 1,177 kilowatts of power, while the supercomputer ranking in the twenty-third spot has a clock speed of 1.035 petaflops using just 493 kilowatts of power. 

Just because the Roadrunner will no longer run nuclear weapons simulations for national security doesn’t mean that it has become completely useless. Scientists at the lab will be studying how the Roadrunner works to inform future supercomputer designs before it is dismantled in about a month. IBM’s Roadrunner will go down in supercomputer history for being the first to reach petaflop speeds and for being a pioneer in processor design. You won’t be forgotten, Roadrunner!

[Image via Los Almos National Laboratory]

Gloria Sin
Former Contributor
Gloria’s tech journey really began when she was studying user centered design in university, and developed a love for…
What happens when AI detectors fail? Researchers say we must be trained to spot fake AI faces
Researchers say spotting AI faces may soon depend more on people than software
Zuckerberg Deepfake

Artificial intelligence has become remarkably good at creating fake human faces. So good, in fact, that the old tricks people relied on - counting fingers, spotting warped earrings, or looking for distorted backgrounds - are quickly becoming obsolete. According to a new study highlighted by the BBC, the next line of defence may not be a better AI detector at all. It might simply be a better-trained human.

Researchers from the University of Aberdeen, working alongside Australia's National University, found that people can dramatically improve their ability to distinguish AI-generated faces from real ones after a relatively short period of structured training. Instead of hunting for obvious visual glitches, participants were taught to recognise subtle patterns that modern image generators still struggle to replicate consistently.

Read more
Google’s new Magic Pointer Play Store listing reveals a Gemini shortcut built for Googlebooks
The unannounced app turns the cursor into a contextual AI tool for search, image creation, and shopping
Plant, Text, Business Card

Google has quietly published a new Play Store listing for Magic Pointer, an unannounced app built for Googlebooks. Updated on July 10, the app turns the cursor into a Gemini shortcut that can act on whatever a user selects on screen.

Magic Pointer can send an image to Lens, generate a related image, or surface a shopping action without forcing users to open a separate chatbot. Regular Android devices currently show as incompatible, so the listing offers an early preview rather than a broad release.

Read more
You can stop using AI, but this new report says you probably can’t escape it
A UK survey found that most people feel AI exposure is unavoidable, raising harder questions about consent, privacy, and whether opting out is still realistic
AI Chatbots

More people are trying to use less AI, but avoiding it altogether may already be impossible.

A survey of 2,055 UK adults found that 42% deliberately limit how much AI they use. Another 70% said avoiding AI exposure would be difficult or impossible, even when they actively wanted less of it.

Read more