• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Tuesday, August 11, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Technology

Scientists Discover Path to Ultra-Low-Energy Data Storage

Bioengineer by Bioengineer
August 1, 2026
in Technology
Reading Time: 3 mins read
0
Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

Artificial intelligence is transforming nearly every corner of modern life, but behind every generated image, internet search, recommendation and scientific simulation lies a rapidly expanding demand for energy. Researchers at the University of Edinburgh have developed a theoretical framework that could help reduce the electricity required to store and manipulate digital information in future magnetic memory technologies. Their approach uses mathematics to design magnetic switching processes that operate with dramatically less energy than conventional methods.

The scale of the challenge is enormous. Data centres already consume vast quantities of electricity to run servers, maintain cooling systems and move information across global networks. As artificial intelligence becomes embedded in healthcare, finance, science, manufacturing and everyday digital services, the amount of data being created and processed is expected to surge. Memory devices, which store the binary states underlying digital information, are a major part of this energy demand because billions or trillions of bits may be switched repeatedly during routine computing operations.

The Edinburgh team focused on magnetic memory, a technology in which information is encoded in the orientation of magnetisation. A magnetic element can represent a binary “0” or “1” depending on whether its magnetic state points in one direction or another. Writing data requires that state to be reversed, a process known as magnetic switching. In current devices, switching is often driven by electrical currents or magnetic fields that are not perfectly tailored to the material’s response, meaning that much of the supplied energy can be dissipated as heat.

To address this inefficiency, the researchers applied Optimal Control Theory, a mathematical method used to determine the most effective way to guide a system from one state to another. Instead of treating a magnetic pulse as a simple on-or-off signal, their framework calculates how the field should vary over time to produce the desired reversal while using as little energy as possible. The model can also incorporate realistic constraints, including limits on pulse strength, switching speed and the physical behaviour of the magnetic material.

In computer simulations, the optimised pulses produced striking results. The calculations suggested that switching energies could be reduced by several orders of magnitude compared with established memory technologies such as dynamic random-access memory, spin-transfer torque magnetic random-access memory and emerging spin-orbit torque magnetic random-access memory. These results do not represent a finished commercial device, but they indicate that the fundamental energy cost of magnetic information processing may be far lower than the cost associated with many existing engineering strategies.

The predicted performance is particularly significant because it approaches the Landauer limit, a fundamental thermodynamic boundary associated with irreversible information processing. The limit states that erasing one bit of information requires a minimum amount of energy proportional to temperature. At room temperature, that minimum is extremely small, but real-world devices operate far above it because of material imperfections, electrical resistance, unwanted heating and control inefficiencies. Moving closer to this limit could make future computing systems substantially more energy efficient, especially when scaled across vast data-centre infrastructures.

The researchers emphasise that the framework is not restricted to magnetic-field pulses. The same mathematical principles could be adapted to switching driven by electrical currents, including the mechanisms used in spintronic devices. It may also be useful for controlling ultrafast laser pulses, which are being investigated for advanced data-storage technologies capable of operating at extraordinary speeds. By changing the control signal while retaining the optimisation strategy, scientists could potentially apply the method to a range of emerging memory architectures rather than a single device design.

Dr Elton Santos of the University of Edinburgh, who led the research, said that every digital operation carries an energy cost and that this cost is becoming increasingly important as artificial intelligence and data-intensive technologies expand. He explained that carefully controlling how a magnetic field changes over time can allow magnetisation to switch much more efficiently than it does under conventional conditions. The work, he added, could represent “the next best thing” for energy-conscious information technology because its underlying mathematics can be transferred to multiple physical systems.

The study, published in Advanced Materials, also offers guidance for turning the theoretical concept into an experimentally testable technology. Future work will need to determine which magnetic materials, device geometries and field-delivery systems can reproduce the simulated performance under laboratory conditions. Researchers will also have to examine how thermal fluctuations, manufacturing imperfections and the need for reliable high-speed operation affect the energy savings. If those obstacles can be overcome, optimised magnetic switching could become an important component of lower-energy computing, helping data infrastructure keep pace with artificial intelligence without allowing its electricity demand to grow unchecked.

Subject of Research: Magnetic memory technologies and energy-efficient information processing

References: Advanced Materials

Image Credits: Dr Elton Santos, University of Edinburgh

Keywords

Artificial intelligence, magnetic memory, magnetic switching, Optimal Control Theory, energy-efficient computing, spintronics, Landauer limit, data centres, information technology, computational modelling

Tags: AI-driven data storage innovationsenergy-efficient data storage solutionsenergy-saving techniques in memory technologyfuture of low-energy data storageimpact of magnetic memory on data center efficiencymagnetic memory for digital informationmagnetic switching process optimizationmathematical modeling of magnetic switchingreducing electricity use in memory devicessustainable data center energy consumptiontheoretical frameworks for magnetic memoryultra-low-energy magnetic memory technology

Share12Tweet7Share2ShareShareShare1

Related Posts

Constructing Digital Twin Brains Beyond Measurement Limits

Constructing Digital Twin Brains Beyond Measurement Limits

August 11, 2026

Brewery-waste biochar could trap harmful bacteria in sandy water filters

August 11, 2026

Computing Professor Wins $584K NSF CAREER Award for Smart-City Research

August 10, 2026

Reference Percentiles for Crying, Regurgitation, Stool Consistency, and CoMiSSTM in Healthy Infants

August 10, 2026

POPULAR NEWS

  • How Cells End Gene Transcription Across Coding and Noncoding Genome Regions

    29 shares
    Share 12 Tweet 7
  • Constructing Digital Twin Brains Beyond Measurement Limits

    29 shares
    Share 12 Tweet 7
  • Brewery-waste biochar could trap harmful bacteria in sandy water filters

    29 shares
    Share 12 Tweet 7
  • Study: Traditional prostate cancer risks don’t predict hormone therapy benefit

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

How Cells End Gene Transcription Across Coding and Noncoding Genome Regions

Constructing Digital Twin Brains Beyond Measurement Limits

Brewery-waste biochar could trap harmful bacteria in sandy water filters

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 86 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.