First processor using light for communications demonstrated

Research teams have developed a single-chip microprocessor that uses light for I/O, by packing two CPU cores with more than 70 million transistors and 850 photonic components onto a 3 x 6-mm chip. They fabricated the device using a standard production foundry. Vladimir Stojanović, an associate professor of electrical engineering and computer science at the Berkeley Wireless Research Center, and fellow UC Berkeley professor Krste Asanović teamed with Rajeev Ram at the Massachusetts Institute of Technology and Miloš Popović at the University of Colorado, Boulder, to develop the new electronic-photonic processor.

The researchers verified the functionality of the chip with the photonic interconnects by using it to run various programs, requiring it to send and receive instructions and data to and from memory. They showed that the chip had a bandwidth density of 300 Gbits/s/mm² , about 10 to 50 times greater than the density of microprocessors currently on the market.

The photonic I/O on the chip is also energy-efficient, using only 1.3 picojoules/bit, equivalent to consuming 1.3 watts of power to transmit a terabit of data per second. In the experiments, the data was sent to a receiver 10 meters away and back.

The electronic-photonic processor chip (Photo credit: Glenn J. Asakawa, University of Colorado).

To enable light to move through the chip with minimal loss, the researchers used the silicon body of the transistor as a waveguide for the light. They did this by using available masks in the fabrication process to manipulate doping. The researchers designed a silicon ring with p-n doped junction spokes next to the silicon waveguide to enable light modulation. Using the silicon-germanium parts of a modern transistor to build a photodetector took advantage of germanium’s ability to absorb light and convert it into electricity. Precise positioning of the fibers over the chip is required.

The new chip is described in a paper in the Dec. 24 print issue of the journal Nature . See the UC Berkeley site at https://goo.gl/f2WSYz.