NYU finds new way to compute
May 13, 2019
and memory devices, such as the hard drives in computers, now use
nanomagnetic mechanisms to store and manipulate information. Unlike
silicon transistors, which have fundamental efficiency limitations, they
require no energy to maintain their magnetic state: Energy is needed
only for reading and writing information.
One method of controlling magnetism uses electrical current that
transports spin to write information, but this usually involves flowing
charge. Because this generates heat and energy loss, the costs can be
enormous, particularly in the case of large server farms or in
applications like artificial intelligence, which require massive amounts
of memory. Spin, however, can be transported without a charge with the
use of a topological insulator - a material whose interior is insulating
but that can support the flow of electrons on its surface.
In a newly published Physical Review Applied paper, researchers from New
York University introduce a voltage-controlled topological spin switch (vTOPSS)
that requires only electric fields, rather than currents, to switch
between two Boolean logic states, greatly reducing the heat generated
and energy used. The team is comprised of Shaloo Rakheja, an assistant
professor of electrical and computer engineering at the NYU Tandon
School of Engineering, and Andrew D. Kent, an NYU professor of physics
and director of the University's Center for Quantum Phenomena, along
Michael E. Flatté, a professor at the University of Iowa.
employs a simple analogy to explain the impact of switching between two
states more effectively. "Imagine if you were preparing a recipe and had
to go into a different room anytime you needed an ingredient before
returning to the kitchen to add it," she says. "It's just as inefficient
when the portions of computing hardware needed to do a calculation and
the portions needed to store it are not well integrated."
While heterostructure devices like theirs, composed of a magnetic
insulator and topological insulator, are still slightly slower than
silicon transistors, vTOPSS increases functionality and circuit design
possibilities, as it has integrated logic and non-volatile memory. "This
is ultimately a matter of user experience and added features," Rakheja
Because vTOPPS will reduce reliance on cloud memory, it also holds the
potential for making computing safer, as hackers will have greater
difficulty gaining access to a system's hardware. Next steps will
include further optimization at the materials and design level to
improve the switching speed, as well as developing prototypes.