VALUE OF INFORMATION-BASED SUSTAINABLE EMBEDDED NANOCOMPUTING
Tackling the emerging hurdles to Moore's law for nanoscale embedded nanocomputing through innovations across computer science, electrical engineering, probability and neurobiology
With the many hurdles facing Moore's law—the maxim that states that the sizes of transistors will halve very two years—at the heart of the electronics and information technology industries, sustaining it past the next decade is viewed as a serious challenge. Increasingly, electronics and computing devices are being used in contexts where human perception is the primary interface to the (embedded) computing engine: cell phones, bio-prosthetics, sensors, signal processing and search technologies are a few examples. The center's guiding philosophy is to take advantage of the limitations in our ability to perceive quality of information from a computer, and when we do perceive it, our willingness to tolerate it if in return, we are able to have access to devices with much lower cost, energy consumption, heat dissipation and an ability to cope with fluctuations in the “quality” of the transistors. Thus, value of information can be used to guide the design of computing devices, while treating many of the impediments as “features” modeled using probability and randomness, with the goal of sustaining the explosive growth of embedded computing in the nanoscale regime, and be increasingly friendly to the energy footprint of computing systems on a global scale.
What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.
—Herbert Alexander Simon