Handling Big Data at Speed of Light
Today’s digital electronic computing is not equipped to efficiently and cost-effectively handle big-data tasks such as weather prediction and biological process modeling. So a multi-university research team led by the University of Arizona is developing new technologies that use optics to increase computing speed and power.
“Digital electronic computing is reaching its limits in cost and capacity,” said UA electrical and computer engineering professor Mark Neifeld, principal investigator on the five-year $7.5 million U.S. Department of Defense Multidisciplinary University Research Initiative, or MURI project. “Optical technology has the potential to improve complex computing in areas ranging from science and health care to business and defense.”
Neifeld, who has a joint appointment in the College of Optical Sciences, and his team will create hybrid computer architectures that combine optical and electronic technologies, allowing for more data to be handled at greater speeds.
Optical computing uses light, or photons, to carry information, and light is faster than electric currents, or electrons, used in conventional computing.
Optics, Neifeld explained, allow for higher bandwidth and massive parallelism, or many processors working in conjunction to perform calculations and program instructions.
Joining Neifeld are UA Optical Sciences professor Nasser Peyghambarian and researchers from the University of California at Berkley, San Diego and Los Angeles.