September 27, 2013
As silicon-based electronics come up against the physical limitations of nanoscale, researchers are scrambling to find a viable replacement that would breath new life into Moore's law and satisfy the demand for ever faster, cheaper and more energy-efficient computers. A new computer made of carbon nanotubes, created by a team of Stanford engineers, may be the first serious silicon challenger.
A scanning electron microscopy image of a section of the first ever carbon nanotube computer. Credit: Butch Colyear
Carbon nanotubes, long chains of carbon atoms, have remarkable material and electronic properties which make them attractive as a potential electronics substrate. The Stanford team, led by Stanford professors Subhasish Mitra and H.-S. Philip Wong, contends that this new semiconductor material holds enormous potential for faster and more energy-efficient computing.
"People have been talking about a new era of carbon nanotube electronics moving beyond silicon," said Mitra, an electrical engineer and computer scientist at Stanford. "But there have been few demonstrations of complete digital systems using this exciting technology. Here is the proof."
According to a paper in the journal Nature, the simple computer is comprised of 142 low-power transistors, each of which contains carbon nanotubes that are about 10 to 200 nanometer long. The prototype has about the same power as a 1970s-era chip, called the Intel 4004, Intel's first microprocessor.
"The system is a functional universal computer, and represents a significant advance in the field of emerging electronic materials," write the authors in the Nature article.
The device employs a simple operating system that is capable of multitasking and can perform four tasks (instruction fetch, data fetch, arithmetic operation and write-back). The inclusion of 20 different instructions from the commercial MIPS instruction set highlights the general nature of this computer. For the demonstration, the team ran counting and integer-sorting workloads simultaneously.
Professor Jan Rabaey, a world expert on electronic circuits and systems at the University of California-Berkeley, noted that carbon had long been a promising candidate to replace silicon, but scientists weren't sure if CNTs would be able to overcome certain hurdles.
While the first carbon nanotube-based transistors came on the scene about 15 years ago, the Stanford team showed that they could be used as the basis for more complex circuits.
"First, they put in place a process for fabricating CNT-based circuits," explained Professor Giovanni De Micheli, director of the Institute of Electrical Engineering at École Polytechnique Fédérale de Lausanne in Switzerland. "Second, they built a simple but effective circuit that shows that computation is doable using CNTs."
By showing that CNTs have a role in designing complex computing systems, other researchers will be more motivated to take the next step, potentially leading to the development of industrial-scale production of carbon nanotube semiconductors.
"There is no question that this will get the attention of researchers in the semiconductor community and entice them to explore how this technology can lead to smaller, more energy-efficient processors in the next decade," observed Rabaey.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?