IBM has announced advances in using light instead of electrical signals to transmit information for future computing, which means Big Data limitations may soon be a thing of the past.
The firm says that it has made a "breakthrough" in technology which exchanges electrical signals for pulses of light in computing -- dubbed "silicon nanophotonics" -- which creates an environment where different optical components can work with electronic circuits on a single silicon chip that uses sub-100nm semiconductor technology.
Silicon nanophotonics aims to make the transfer of vast quantities of data quicker through computer chips contained in servers, large datacenters, and supercomputers. By using pulses of light instead of electronic signals, the idea behind the research field is to alleviate congestion and traffic between computer relay points -- as well as cutting down the expense of traditional interconnection.
The amount of data we produce, contain and store is growing. Dubbed "Big Data", storing and analyzing such information is now cheap enough that many enterprises can take advantage of information in order to predict consumer trends, gain a more insightful, and also keep track of businesses themselves. However, with the 'data explosion', there is a growing demand for better chip performance and higher levels of computing power.
"This technology breakthrough is a result of more than a decade of pioneering research at IBM," said Dr. John E. Kelly, Senior Vice President and Director of IBM Research. "This allows us to move silicon nanophotonics technology into a real-world manufacturing environment that will have impact across a range of applications."
Building on a proof of concept developed in 2010, the firm has added variants of silicon components including wavelength division multiplexers (WDM), modulators, and detectors into a side-by-side CMOS electrical circuitry board as part of a 90nm CMOS fabrication line. By doing so, IBM hopes that silicon nanophotonics technology can become commercially viable.
The transceivers were able to exceed a data rate of 25Gbps per channel. In addition, the technology was capable of coping with a number of parallel optical data streams, twisting them into a single fiber by utilizing compact on-chip wavelength-division multiplexing devices. In turn, the idea of multiplexing large data streams at high rates -- coping with the demands of Big Data -- may soon be on the commercial horizon.
Image credit: Dennis Skley
This post was originally published on Smartplanet.com