Electronic Information Processing After Scaling: Physical Limits for Technology Assessment

摘要：随着硅集成电路技术的规模化接近收益递减点，以及后续技术的建议激增，对新的和非常规的信息处理方法进行批判性评估的需求将越来越大。我们需要新的工具来衡量潜在的后续技术的最终潜力，以便在速度和电路密度上可靠地处理信息，而这些信息的处理速度和电路密度将超过那些在硅中实现的扩展。该计划旨在通过扩展和应用物理信息理论来开发这些工具，将这一理论方法的力量应用于新兴电子信息处理技术的实际评估。在技术相关的环境中，对信号保真度的基本限制和信息处理的物理成本（如散热）的探索将得到最高的优先考虑。这个项目的成功将为评估新兴技术（包括但不限于基于电荷的纳米电子学）的信息处理能力和资源需求提供有用的方法，并将促进对信息处理的物理方面的理解和对其实际后果的欣赏。

CCF - 0649691 Electronic Information Processing After Scaling: Physical Limits for Technology AssessmentPI: Anderson, Neal G. Inst: University of Massachusetts AmherstAbstract:As scaling of silicon integrated circuit technology approaches the point of diminishing returns, and proposals for successor technologies proliferate, there will be a growing need for critical assessment of new and unconventional approaches to information processing. New tools will be needed to gauge the ultimate potential of would-be successor technologies to reliably process information at speeds and circuit densities exceeding those that will have been achieved in silicon at the end of scaling. This program aims to develop such tools through extension and application of physical information theory, bringing the power of this theoretical approach to bear on the practical assessment of emerging electronic information processing technologies. Exploration of fundamental limits on signal fidelity and the physical costs of information processing (e.g. in heat dissipation) in technologically relevant settings will receive the highest priority. Success in this program will provide useful approaches for evaluating the information processing capabilities and resource requirements of emerging technologies - including but not limited to charge-based nanoelectronics - and will promote understanding of the physical aspects of information processing and appreciation of their practical consequences.