Professional Documents
Culture Documents
Computing at The Speed of Light - MIT Technology Review
Computing at The Speed of Light - MIT Technology Review
Intelligent Machines
“All communications over long distance
are driven by lasers, but you’ve never had
it inside devices,” says Mario Paniccia,
director of Intel’s photonics lab in Santa
Clara, CA. “Our new integrated optical
link makes that possible.”
Paniccia’s team has perfected tiny silicon
chips capable of encoding and decoding
laser signals sent via fiber optics. Today,
Seeing the light: A chip in the center of this circuit board
when data arrives at a computer via a fiber contains four lasers that convert electrical signals into light
optic connection it has to be moved from a pulses. The pulses travel at high speeds along a fiber-optic
link.
separate photonic device to an electronic
circuit. This new system promises to speed things up because everything
works in silicon.
Last week, Paniccia’s team demonstrated the first complete photonic
communications system made from components fully integrated into
silicon chips. Electronic data piped into one chip is converted into laser
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 1/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
light that travels down an optical fiber and is transferred back into
electrical signals a few fractions of a second later. The system can carry
data at a rate of 50 gigabits per second, enough to transfer a fulllength
HD movie in less than a second.
The silicon photonic chips could replace the electronic connections
between a computer’s key components, such as its processors and
memory. Copper wiring used today can carry data signals at little more
than 10 gigabits per second. That means critical components like the
central processing unit and the memory in a server cannot be too far
apart, which restricts how computers can be built.
The new Intel setup has four lasers built into its transmitter chip that
shine data into a single optical fiber at slightly diûerent wavelengths, or
“colors.” Chips with even more lasers should make it possible to
communicate at 1,000 gigabits per second.
“Having a chip the size of your fingernail that can deliver a terabit per
second changes the way you can think about design,” says Paniccia. Such
chips could make a big diûerence inside the sprawling data centers
operated at great expense by Web giants like Google, Microsoft, and
Facebook. “Data centers today are big piles of copper–that imposes the
limits on how you arrange components inside a server,” Paniccia says.
“If I could just move the memory a foot away [from the processors], I
could add a whole board of memory for a single CPU instead,” says
Paniccia, whose team is experimenting with prototype servers to work
out how to build them with photonics links inside.
Moving a server’s memory away from the CPUs would also make
ventilating them easier. Since roughly half the cost of running a data
center, used for everything from services like Facebook to banking
records, comes from cooling, that could have a significant impact.
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 2/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
Further savings may come from the fact that optical links require less
power to operate, says Keren Bergman, who leads a silicon photonics
research group at Columbia University. “With electrical wires, the longer
you go, the more energy you spend in an exponential fashion,” she says.
Optical fiber allows lowpower signals to travel farther faster. Bergman’s
group has used data on the performance of computers at Lawrence
Berkeley and MIT’s Lincoln Laboratories to simulate how systems with
optical interconnects might perform. “You can get an order of
magnitude gain in energy eùciency,” she says, with the largest gains
seen for applications such as highbandwidth image processing and
video streaming, she says.
Data centers aren’t the only things that may see their insides lit up with
lasers. “We’ve developed this technology to be lowcost so we can take it
everywhere, not just into highperformance computing or the data
center,” says Paniccia. The components of the Intel system, including the
lasers, are made with the same siliconsculpting methods used to
construct computer chips in vast quantities. “I’m drafting Moore’s law,”
says Paniccia. “We’ve enabled the benefits of using light with the low
cost, highvolume, scalability of silicon.”In consumer computers like
laptops, that would allow innovations in industrial design. I could put
the memory in the display instead, and change the design of the whole
thing.”
This could make it easier to swap in new components without having to
open up a machine. It would also allow core components to be installed
in peripherals. Extra memory could, for example, be hidden in a laptop
or smart phone dock to increase a portable device’s computing power
when plugged in.
Fully exploiting the benefits of the optical age will, however, means
changes to the components being linked up. “It’s not just a case of whip
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 3/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
out the electrical wires and replace them with optical fiber,” says
Bergman.
Ajay Joshi, an assistant professor at Boston University, who is also
exploring design options for highperformance computers with optical
interconnects, agrees. “If we speed up the channel between logic
[processors] and memory, we need to rethink the way you design that
memory.”
The speed gap between processors and optical links is smaller, but
ultimately, that too will likely change. “It would be nice to also see
processors that work optically instead of electronically,” Joshi says.
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 5/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
Subscribe
See details+
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 6/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 7/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 8/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 9/10
3/30/2017 Computing at the Speed of Light MIT Technology Review
https://www.technologyreview.com/s/420082/computingatthespeedoflight/ 10/10