What innovations are possible with ultra-efficient sensors?

What innovations are possible with ultra-efficient sensors?

If you think we're in the age of Big Data now, just wait: It's only just begun.

On stage at VERGE DC this morning, Jon Koomey walked through the steady progression of energy efficiency in computing devices -- the energy corollary to Moore's Law -- and what it means for innovations in how we use IT for sustainability.

Koomey, a consulting professor at Stanford University and one of the leading researchers on energy-efficient computing (and author of a recent book on energy and climate innovation), explained the widespread understanding of Moore's Law, namely that the performance of microprocessors is doubling every 1.5 years.

"One thing people don't know about is that the efficiency of computing in that same period, from 1975 to today, has also doubled every 1.5 years," Koomey said. "And that trend started in the 1940s, it started in the era of vacuum tubes, before the microprocesser was a glint in someone's eye."

The ever-more-efficient operation of computing has already changed the world: Koomey said that smartphones and laptops, along with countless other technologies, owe their existence to the twin trends of increased performance and increased efficiency.

And the evolution of mobile computing devices still has a long way to go. Koomey described how Josh Smith at the University of Washington has created a device that can scavenge all the power it needs -- about 60 microwatts -- from ambient television and radio signals. And more and more researchers are working on "smart dust" technologies, which are, as you'd expect, clouds of sensors arrayed across an environment.

Koomey said that this evolution of technology will vastly improve the kinds of data we can collect in our sensors -- "If you think you've seen Big Data now, you ain't seen nothing yet." -- but that rather than looking big, we should start looking small.

"Instead of focusing on big data, we should, as Erik Byrnolfsson at MIT says,  start looking at nanodata; data about individuals and transactions that when analyzed in the right way can actually lead to great insight into new ways to wrest value from the environment."

This will take any number of shapes, whether from extremely fine-tuned control, in real-time, over processes and devices or matching energy supply very precisely to demand from devices at any scale -- or in other words, significantly less wasted energy from devices.

In closing, Koomey offered a challenge to the entrepreneurs and innovators in the room at VERGE:

In keeping with Alan Keyes' dictum that the best way to predict the future is to invent it, what innovations can you imagine, building on the steady trends in efficiency and performance that are leading to an explosion in the use of mobile sensors, computing and controls, that will radically transform your ability to accomplish your goals, and then what concrete steps are you going to take to make those innovations a reality?

Microprocessor photo via Shutterstock.