CITRIS Signal Newsletter: Bits Add Up

Dear Friends of CITRIS:

CITRIS is a high-performance, high-energy organization, and we are pleased to feature two projects demonstrating this theme. The first confirms a new phenomenon observed by researchers in the Marvell NanoLab, one with exciting implications for the semiconductor industry. Warnings that we have nearly exhausted the exponential ride fueled by Moore’s Law (which says that the number of transistors in an integrated circuit will double approximately every two years) have caused mounting alarm in industry and academia for years. For one thing, unless we can continue to make our proliferating devices less power-hungry, they will demand ever increasing portions of our shared energy resources. For another, the technological (and to a large degree economic) progress mapped out for the coming century is largely premised on ever smaller, cleaner, faster and more efficient processors.

UC Berkeley professors Sayeef Salahuddin and Ramamoorthy Ramesh have just published a promising paper in Nature Materials that may help extend Moore’s trajectory for some time to come. For the first time, they have directly witnessed a phenomenon called negative capacitance wherein the voltage of a material goes down when a charge is applied. That counterintuitive response could allow engineers to dip below the current standard for the minimum charge required to store one bit of information. Multiply that savings by the billions of transistors in a single laptop and multiply that by billions of laptops, phones, tablets, and other digital devices under development—the bits really add up.

We are delighted (though not surprised!) that such groundbreaking research occurred within the CITRIS ecosystem, including the Marvell NanoLab. As our industry partners continue to innovate and push the limits of Moore’s upward curve within the current paradigm, our academic mandate is to look beyond that paradigm into the future. This discovery demonstrates that such fundamental high-risk, high-reward work is exactly what we should be investing in. Read our first story, “Finding Hidden Joules”.

Our second story, “Managing Local Power,” describes another angle on accomplishing more with less energy. Funded in part by the CITRIS seed grant program, researchers from different UC campuses and disciplines are collaborating on ways to make micro-grids fed by renewable sources responsive to fluctuations in the energy supply. If a small local grid runs largely on solar, for example, and available energy dips suddenly when a cloud covers the sun, the system must do something to compensate. It can either start drawing from the macro-grid, causing a spike in demand, or it can temporarily reduce its own use by turning down (or off) non-essential appliances. Unpredictably hitting the macro-grid with demand spikes is expensive, inefficient, and destabilizing. But adjusting localized demand in real time can be tricky. This multi-campus project at UC Davis and UC Santa Cruz examines some of the technical challenges in microgrid management. Equally important, it examines why people make the energy-use decisions they do. Understanding these behavioral motivations is a precondition for nudging people toward change, which may include relinquishment of some individual choices and actions to management by a smart micro-grid. This project also involves an exchange program with Aalborg University and the Technical University of Denmark, reflecting our growing engagement in international partnerships here at CITRIS.

We wish you all the best for the holidays and an inspiring new year.

costas_spanosCostas Spanos
Director, CITRIS and the Banatao Institute@CITRIS Berkeley

&

Camille_Crittenden_headshot_2014Camille Crittenden
Deputy Director, CITRIS and the Banatao Institute@CITRIS Berkeley