Charge of the Light Brigade: Faster and More Efficient New Chips Using Photons Instead of Electrons

"PACE - PEACE" Image by Etienne Valois

“PACE – PEACE” Image by Etienne Valois

Alfred, Lord Tennyson wrote his immortal classic poem, The Charge of the Light Brigade, in 1854. It was to honor the dead heroes of a doomed infantry charge at the Battle of Balaclava during the Crimean War. Moreover, it strikingly portrayed the horrors of war. In just six short verses, he created a monumental work that has endured ever since for 162 years.

The poem came to mind last week after reading two recent articles on seemingly disparate topics. The first was posted on The New Yorker’s website on December 30, 2015 entitled In Silicon Valley Now, It’s Almost Always Winner Takes All by Om Malik. This is highly insightful analysis of how and why tech giants such as Google in search, Facebook in social networking, and Uber in transportation, have come to dominate their markets. In essence, competition is a fierce and relentless battle in the global digital economy. The second was an article on posted on December 23, 2015 entitled Chip Promises Faster Computing with Light, Not Electrical Wires by Stephan Shankland. I highly recommend reading both of them in their entirety.

Taken together, the homonym of “light” both in historical poetry and in tech, seems to tie these two posted pieces together insofar as contemporary competition in tech markets is often described in military terms and metaphors. Focusing on that second story here for purposes of this blog post, about a tantalizing advance in chip design and fabrication, will this survive as it moves forward into the brutal and relentlessly “winner takes all” marketplace? I will summarize and annotate this story, and pose some of my own, hopefully en-light-ening questions.

Forward, the Light Brigade

A team of researchers, all of whom are university professors, including Vladimir Stojanovic from the University of California at Berkeley who led the development, Krste Asanovic also from Berkeley, Rajeev Ram from MIT, and Milos Popovic from the University of Colorado at Boulder, have created a new type of processing chip “that transmits data with light”. As well, its architecture significantly increases processing speed while reducing power consumption.  A report on the team’s work was published in an article in the December 24, 2015 issue of Nature (subscription required) entitled Single-chip Microprocessor That Communicates Directly Using Light by Chen Sun, Mark T. Wade, Yunsup Lee, et al.

This approach, according to Wikipedia, of “using silicon as an optical medium”, is called silicon photonics. IBM (see this link) and Intel (see this link)  have likewise been involved in R&D in this field, but have yet to introduce anything ready for the market.

However, this team of university researchers believes their new approach might be introduced commercially within a year. While their efforts do not make chips run faster per se, the photonic elements “keep chips supplied with data” which avoids them having to lose time by idling. Thus, they can process data faster.

Currently (no pun intended), electrical signals traverse metal wiring across the world on computing and communications devices and networks. For data traveling greater national and international distances, the electronic signals are transposed into light and sent along on high-speed fiber-optic cables. Nonetheless, this approach “isn’t cheap”.

Half a League Onward

What the university researchers’ team has done is create chips with “photonic components” built into them. If they succeed in scaling-up and commercializing their creation, consumers will be likely the beneficiaries. These advantages will probably manifest themselves first when used in data centers that, in turn, could speed up:

  • Google searches
  • Facebook image recognition
  • Other “performance-intensive features not economical today”
  • Remove processing bottlenecks and conserve battery life in smartphones and other personal computing platforms

Professor Stojanovic believes that one of their largest challenges if is to make this technology affordable before it can be later implemented in consumer level computing and communications devices. He is sanguine that such economies of scale can be reached. He anticipates further applications of this technology to enable chips’ onboard processing and memory components to communicate directly with each other.

Additional integrations of silicon photonics might be seen in the lidar remote sensing systems for self-driving cars¹, as well as brain imaging² and environmental sensors. It also holds the potential to alter the traditional methods that computers are assembled. For example, the length of cables is limited to the extent that data can pass through them quickly and efficiently before needed amplification along the way. Optical links may permit data to be transferred significant further along network cabling. The research team’s “prototype used 10-meter optical links”, but Professor Stojanovic believes this could eventually be lengthened to a kilometer. This could potentially result in meaningful savings in energy, hardware and processing efficiency.

Two startups that are also presently working in the silicon photonics space include:

My Questions:

  • Might another one of silicon photonics’ virtues be that it is partially fabricated from more sustainable materials, primarily silicon derived from sand rather than various metals?
  • Could silicon photonics chips and architectures be a solution to the very significant computing needs of the virtual reality (VR) and augmented reality (AR) systems that will be coming onto the market in 2016? This issue was raised in a most interesting article posted on on December 30, 2015 entitled Few Computers Are Powerful Enough to Support Virtual Reality by Ian King. (See also these 13 Subway Fold posts on a range of VR and AR developments.)
  • What other new markets, technologies and opportunities for entrepreneurs and researchers might emerge if the university research team’s chips achieve their intended goals and succeed in making it to market?

May 17, 2017 UpdateFor an update on one of the latest developments in photonics with potential applications in advanced computing and materials science, see Photonic Hypercrystals Are Now a Reality and Light Will Never Be the Same, by Dexter Johnson, posted on May 10, 2017, on 

1.  See these six Subway Fold posts for references to autonomous cars.

2.  See these four Subway Fold posts concerning certain developments in brain imaging technology.

The Next Wave in High Tech Materials Science


Optical Profilometer Metamaterials, Image by Oak Ridge National Laboratory

Metamaterials are not something used by the United Federation of Planets’ engineers to build the next iteration of the Starship Enterprise (which, btw, would be designated the NCC-1701-F, although some may differ).  Rather, they are materials fabricated in such a manner that they can bend light, sound, radar, radio and seismic waves. The technological implication of applying these materials in antennas, radar, cosmetics and soundproofing may prove to be transformative according to a fascinating article in the March 23, 2015 edition of The New York Times entitled The Waves of the Future May Bend Around Metamaterials, by John Markoff.  I will summarize this, add some links and annotations, and pose some questions.

These substances achieve their remarkable effect by being composed of microscopic “subcomponents” that are smaller than the wavelengths of the types of waves they are engineered to bend in certain ways. That is, they can be used to “manipulate” the waves in designated manners “that do not normally occur”.

Researchers have been developing a variety of metamaterials for the past 15 years. Their work has recently begun yielding some genuine innovations in systems that incorporate these advances in original and innovative ways. Some of these latest developments include:

  • Airbus*and Lamda Guard are about to test a coating on airline windows to deter attempts to blind them with laser pointing devices by someone on the ground. (See NYC Man Charged With Pointing Laser at Aircraft, in the March 15, 2015 edition of The New York Times for a recent case of this here in New York.)
  • Echodyne is working on several types of antennas, radar-based navigation systems and other devices.
  • Evolv Technology is developing airport security systems.
  • Kymeta has partnered with Intelsat to engineer “land-based and satellite-based intelligent antennas”.
  • Dr. Xiang Zhang at the University of California at Berkeley, is working on, among other metamaterials projects, “superlenses” for microscopes that might increase their magnification powers beyond today’s capabilities. He has received inquiries from “military contractors and commercial companies” and even cosmetics companies concerning metamaterials. As well, he and other developers are creating apps for optical computer networks.
  • Professor Vinod Menon and his research team at the City College of New York, in their Laboratory for Nano and Micro Photonics, have demo-ed “light emission from ultrafast-switching LEDs” made from metamaterials. Using this and other related developments may also lead to significantly faster optical computers networks.
  • Menard Construction published a paper in 2013 entitled Seismic Metamaterial: How to Shake Friends and Influence Waves? by S. Brûlé, E.H. Javelaud, S. Enoch and S. Guenneau, where the company successfully tested “a metamaterial grid of empty cylindrical columns bored into soil” in an effort to reduce the effects of a “simulated earthquake”. (The phases in quotes in the last sentence were from the NYTimes article, not the research paper itself.)

The article concludes on a note of great optimism from Professor Zhang about the future of metamaterials. I completely agree. Once these apps and development projects make their way into commercial markets and other scientists and companies from different fields and industries take greater notice, I strongly believe that new forms of metamaterials and their applications will emerge that have not even been imagined yet. Like any dramatically new technology, this will find its applications perhaps in some very unlikely and surprising sectors.

Just to start off, what about medical devices, optical computing and storage devices, visual displays, sound and video recording, and automotive safety technology? Let’s keep watching and see what springs from people’s needs and creativity.

Finally, just a quick mention of a recently published book that received many excellent reviews for a lively and engaging series of stories about the key developments of basic materials and materials science through history entitled Stuff Matters: Exploring the Marvelous Materials That Shape Our Man-Made World by Mark Midownik (Houghton Mifflin Harcourt, 2014).

[While I hope that this blog post will be enlightening, please be assured that no light waves were bent or harmed during the drafting process.]


Another innovative project by Airbus to develop a drone for bringing Net access to remote and under-served regions was covered in the November 26, 2014 Subway Fold post entitled Robots and Diamonds and Drones, Aha! Innovations on the Horizon for 2015.