I will review physics of the so-called "thermodynamic limit" on the energy consumption at computation, and C.Bennett's idea of reversible computing, which allows that limit to be avoided. Unfortunately, even if implemented in hardware virtually free of static power consumption (such as Parametric Quantron circuits), a genuinely reversible computation would require exponentially large resources. Selective reversibility sacrifices may sharply reduce this hardware overhead, but still leave the circuit speed and defect tolerance relatively low.

Amory Lovins is widely considered among the world's leading authorities on energy, especially its efficient use and sustainable supply.  As Chairman and Chief Scientist of the Rocky Mountain Institute, Lovins pioneered the concept of "soft energy paths" involving efficient energy use, diverse and renewable energy sources, with special reliance on "soft energy technologies" such as solar, wind, biofuels, geothermal, etc., matched in scale and quality to their task, and widely accessible across society.

Interest has been growing in powering data centers (at least partially) with renewable or "green" sources of energy, such as solar or wind.  However, it is challenging to use these sources because, unlike the "brown" (carbon-intensive) energy drawn from the electrical grid, they are not always available.  In this talk, I will first present a case for leveraging green energy in small and medium-scale data centers.  I will then describe the solar-powered data center we have just built at Rutgers, and the lessons we have learned building it.

Big first order phase transformations in solids can still be highly reversible, if the lattice parameters are “tuned” to satisfy certain relations that promote the compatibility between phases.  We present recent measurements of hysteresis in martensitic materials resulting from this kind of tuning.  We re-examine the origins of hysteresis in light of these measurements, and conclude that a certain energy barrier, not pinning or thermal activation, is primarily responsible for hysteresis in a broad array of materials that undergo phase transformations.

We have demonstrated efficient solution-processed small-molecule solar cells based on a novel molecular donor, DTS(PTTh2)2. A record power conversion efficiency (PCE) is achieved for small-molecule organic photovoltaics: PCE=7% under AM 1.5 G irradiation (100 mW cm–2) from bulk heterojunction (BHJ) composites of DTS(PTTh2)2:PC70BM (donor to acceptor ratio of 7:3) with short circuit current (Jsc) of 14.4 mA cm–2, open circuit voltage (Voc) of 0.78 V and fill factor (FF) of 59%.

More than ten years ago, it was envisioned that the interconnects will be the limiters for continued increase in compute performance. Now we know that it's not the interconnects but power and energy that has been the limiter. As technology scaling continues providing abundance of transistors, and new architectures to continue to deliver performance in a given power envelope, we need to revisit the role of interconnects.

One of the greatest challenges of the 21st century will be to understand, invent, and engineer new mechanisms and materials for energy production, energy storage and energy transport to counter the deleterious environmental and political impacts of our long-standing reliance on crude oil.  Current renewable energy conversion and storage technologies are too expensive, too inefficient, or both, substantially limiting their use and global impact.

Transition metal oxides exhibit extraordinary phenomena not observed in any other materials class.  For example, they show “Mott” metal-insulator transitions (MITs) that are caused by strong electron correlations.  While properties have been studied extensively in bulk materials, oxide heterostructures allow for new approaches to control their properties and to study the fundamental physics, through the manipulation of carrier concentration by composition, modulation doping, field effect, two-dimensionality and quantum confinement, which cannot be realized with bulk materials.

Life in the time of Dennard scaling was relatively easy for architects and the computer industry. Every process generation delivered twice as many transistors to a chip that could run at a 1.4 times faster clock rate and consume the same power as the previous generation. General purpose processors spent this bounty on deep pipelining for high clock rates, extreme out-of-order execution to mine instruction-level parallelism, and large on-chip caches.

World Governments cannot sustainably fund uneconomical schemes of incentives and global projects aimed at slowing the acceleration of atmospheric carbon dioxide levels. Not even at maximal implementation would wind, solar, tidal energy, low emission vehicles, carbon capture and sequestration, and further emission controls on carbon-based electrical power generation and industrial plants have a significant near-term impact.