The magnet-free computer of the future 26 August 2014

Meet the magnet-free computer of the future

    Israeli research closes in on the possibility of an inexpensive, energy-efficient computer as small as a flash drive and as powerful as a mainframe.
  • icon_zoom.png
    Prof. Yossi Paltiel, leading the way to tomorrow’s tiny computer Prof. Yossi Paltiel, leading the way to tomorrow’s tiny computer
    By Rivka Borochov 
    Israelis historically have taken a leading role in amping up the performance and storage capacity of computers.
    Researchers from Tel Aviv University invented the flash drive –– or disk-on-key–– and now researchers at the Hebrew University and the Weizmann Institute of Science may take memory storage from the edge of impossible to the next level. 
    The latest research from Israel closes in on the possibility of making a portable computer as small as a flash drive, as powerful as a mainframe, at a very reasonable cost, using few power resources. The very idea pushes the boundaries of technology. 
    Led by Prof. Yossi Paltiel from the Hebrew University’s School of Engineering and Computer Science, the scientists have found a way to eliminate the need for permanent magnets in memory devices. This opens the door to many new innovative computing technologies, they say.  
    Their new approach, magnetless spin memory (MSM), could be the watchword for computer engineers of the future. “The magnetless spin memory has the potential to become the basis of a whole new generation of faster, smaller and less expensive memory technologies,” says Paltiel. 
    He explains that memory devices such as hard drives, flash drives and RAM are the core of our computers, phones, cars and electronic appliances. But dynamic RAM memory has to be refreshed continually, and can be lost in the event of a power surge or outage. Flash memory lacks speed. And all of these technologies are limited by size. 
    “There are people working on magnetic memory that performs at a small scale, but the materials they are using to write with are complicated, expensive and hard to miniaturize,” says Paltiel. 
    But the trend, or race, in computing science is overcoming this performance bottleneck to achieve high speeds that are attainable and useful for everyday people and the devices they are buying at ever cheaper prices. 
    The bottleneck is overcome using quantum physics.  
    “Here in our new research, we don’t use magnets, but apply a magnetized current to lay down new memory. Approaching the nanoscale level, we are using a chiral molecule, one that can magnetize a current so that we can write a magnetic beat,” Paltiel explains. 
    Laptops in a flash 
    “Our research is to pave the way for dreams of universal memory -- memory that is uncompromisingly small, fast, very cheap and has near unlimited potential.  
    “A flash drive as a computer? It’s theoretically possible. That’s what we are dreaming about,” he says. 
    The Israeli research, published in Nature Communications, hinges around how electrons “spin” in memory devices. Currently, layers of memory are etched into hard drives with the help of heavy, expensive magnets. Paltiel and his associates can magnetize at a tiny, quantum scale – without actual magnets -- by influencing the spin flow of electrons to create a current that transfers memory to nano-sized particles.  
    “We haven’t closed any deals yet, but the technology holds a lot of promise,” says Paltiel, explaining that the process must still be fine-tuned to operate under higher room temperatures. 
    The research is already drawing international interest for its commercial potential and is currently being developed for commercialization by the tech transfer companies Yissum and Yeda, of the Hebrew University and the Weizmann Institute, respectively. 
    After a significant investment is made in the basic research, Paltiel assumes that it could take two years to demonstrate the technology has commercial potential and several years more for it to become available. 
    “Our initial quest is for a simple memory device –– one which is fast in computing power and as long-lasting as a disk-on-key or a hard drive that we know today,” he says. “It’s basic research, after all, and that’s what we do and are excited about. We would be happy to take this research to a startup, though, and to collaborate with those who see the enormous applications like we do.”