HPC in the Cloud: IBM Researchers Create World’s Smallest Magnetic Memory Bit

Dedicated to covering high-end cloud computing
in science, industry and the datacenter

Translation Disclaimer

Subscribe | Sign In

January 12, 2012

IBM Researchers Create World’s Smallest Magnetic Memory Bit

New experimental atomic-scale magnet memory is at least 100 times denser than today’s hard disk drives and solid state memory chips.

SAN JOSE, Calif., Jan. 12 -- Punctuating 30 years of nanotechnology research, scientists from IBM Research have successfully demonstrated the ability to store information in as few as 12 magnetic atoms. This is significantly less than today’s disk drives, which use about one million atoms to store a single bit of information. The ability to manipulate matter by its most basic components – atom by atom – could lead to the vital understanding necessary to build smaller, faster and more energy-efficient devices.

While silicon transistor technology has become cheaper, denser and more efficient, fundamental physical limitations suggest this path of conventional scaling is unsustainable. Alternative approaches are needed to continue the rapid pace of computing innovation.

By taking a novel approach and beginning at the smallest unit of data storage, the atom, scientists demonstrated magnetic storage that is at least 100 times denser than today’s hard disk drives and solid state memory chips. Future applications of nanostructures built one atom at a time, and that apply an unconventional form of magnetism called antiferromagnetism, could allow people and businesses to store 100 times more information in the same space.

“The chip industry will continue its pursuit of incremental scaling in semiconductor technology but, as components continue to shrink, the march continues to the inevitable end point: the atom. We’re taking the opposite approach and starting with the smallest unit -- single atoms -- to build computing devices one atom at a time.” said Andreas Heinrich, the lead investigator into atomic storage at IBM Research – Almaden, in California.

The research was published today in the peer-reviewed journal Science.

How it Works

The most basic piece of information that a computer understands is a bit. Much like a light that can be switched on or off, a bit can have only one of two values: "1" or "0". Until now, it was unknown how many atoms it would take to build a reliable magnetic memory bit.

With properties similar to those of magnets on a refrigerator, ferromagnets use a magnetic interaction between its constituent atoms that align all their spins – the origin of the atoms’ magnetism – in a single direction. Ferromagnets have worked well for magnetic data storage but a major obstacle for miniaturizing this down to atomic dimensions is the interaction of neighboring bits with each other. The magnetization of one magnetic bit can strongly affect that of its neighbor as a result of its magnetic field. Harnessing magnetic bits at the atomic scale to hold information or perform useful computing operations requires precise control of the interactions between the bits.

The scientists at IBM Research used a scanning tunneling microscope (STM) to atomically engineer a grouping of twelve antiferromagnetically coupled atoms that stored a bit of data for hours at low temperatures. Taking advantage of their inherent alternating magnetic spin directions, they demonstrated the ability to pack adjacent magnetic bits much closer together than was previously possible. This greatly increased the magnetic storage density without disrupting the state of neighboring bits.

Writing and reading a magnetic byte: this image shows a magnetic byte imaged 5 times in different magnetic states to store the ASCII code for each letter of the word THINK, a corporate mantra used by IBM since 1914. The team achieved this using 96 iron atoms − one bit was stored by 12 atoms and there are eight bits in each byte.

IBM and Nanotechnology Leadership

In the company's 100 year history, IBM has invested in scientific research to shape the future of computing. Today's announcement is a demonstration of the results garnered by IBM's world-leading scientists and the company's continual investment in and focus on exploratory research.

IBM Research has long been a leader in studying the properties of materials important to the information technology industry. For more than fifty years, scientists at IBM Research have laid the foundation of scientific knowledge that will be important for the future of IT and sought out discoveries that can advance existing technologies.

For additional information, including high-resolution images and an animation explaining the technique, visit www.ibm.com/atomicscalememory.

-----

Source: IBM

Share Options

Subscribe

Subscribe to HPC in the Cloud

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Content Type

Subscribe to All Content

All

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

AWS, SAP Announce Certification of AWS Infrastructure for SAP Business All-in-One Solutions

May 10, 2012

May 09, 2012

May 08, 2012

May 07, 2012

Feature Articles

Cloud Services Satisfy a Higher Calling

Higher education involves many collaborative projects that lend themselves to cloud services, however often those services are not tailored to the uniqueness of an academic environment. That's where the Internet2 NET+ project comes in. By partnering with 16 major cloud providers, the networking consortium is seeking to expedite the delivery of cloud services and by doing so advance research and innovation in the United States.
Read more...

Univa CEO Gary Tyreman on the Evolution of HPC, Big Data and Cloud

It's been a little over a year since Univa took over stewardship of the open source workload manager and acquired the founding Sun Grid Engine team from Oracle, and the company just announced its third production release. CEO Gary Tyreman discusses the latest enhancements as well as the company's plans around cloud, big data and the enterprise.
Read more...

Read more HPC in the Cloud features...

Around the Web

Breaking the Cloud Barrier

May 15, 2012 | New Microsoft report shows that beyond the expected financial benefits, cloud services may offer more comprehensive security features compared to in-house IT operations.
Read more...

Vendors Demo Next-Gen Sequencing Platforms for Pharma

May 14, 2012 | During the second annual Pistoia Alliance conference, three teams demonstrated their newly-implemented cloud-based next-generation sequencing platforms.
Read more...

Zunicore Offers Bare Metal by the Hour

May 10, 2012 | PEER1's cloud division, Zunicore, will soon be offering GPU-equipped servers on-demand.
Read more...

US Cloud Providers Struggle With Data Privacy Laws

May 08, 2012 | The Patriot Act leads foreign governments to question the security of US cloud services.
Read more...

Genomic Data Gets Comfy in the Cloud

May 07, 2012 | Startup DNANexus is helping to bring personalized medicine to the masses with its cloud-based next-gen sequencing platform.
Read more...

Read more headlines...

Appro White Paper: Enabling Performance-per-Watt Gains in HPC

04/05/2012 | Appro | Designed to meet the growing global demand for HPC solutions, Appro's Xtreme-X™ Supercomputer delivers superior performance-per-watt and reduced I/O latency while bringing significant flexibility to HPC workload configurations including capacity, hybrid, data intensive and capability computing.

Exploring the Potential of Heterogeneous Computing

04/02/2012 | AMD | Developers today are just beginning to explore the potential of heterogeneous computing, but the potential for this new paradigm is huge. This brief article reviews how the technology might impact a range of application development areas, including client experiences and cloud-based data management. As platforms like OpenCL continue to evolve, the benefits of heterogeneous computing will become even more accessible. Use this quick article to jump-start your own thinking on heterogeneous computing.