PFI:AIR - TT: Scalable Accelerated Power Simulation Program with Integrated Circuit Emphasis

这个PFI：AIR技术翻译项目致力于将一种新型集成电路(IC)功率估计仿真技术转化为一种优于当前最先进水平的集成仿真工具。以集成电路为重点的可扩展加速功率模拟计划(SAP-SPICE)将显著提高功率估计模拟的速度和准确性，从而极大地改善IC制造商设计、制造和测试现代芯片的方式。该项目解决了以下技术差距，因为它从研究发现转化为商业应用。虽然黄金标准的模拟工具HSPICE产生了最准确的结果，但它的计算成本很高。另一方面，加速模拟器速度更快，因为它们依赖于更大的组件模型，但它们产生的平均结果不太准确。SAP-SPICE项目使用一种不同的方法进行逻辑设计、逻辑模拟器和部分功率数据，以在保持快速计算速度的同时提取更准确的结果。该项目的目标是生产一个试生产质量的原型，该原型将与现有的行业标准工具集成，向潜在客户展示该技术相对于其他工具的优势。SAP-SPICE将在几个方面产生更广泛的影响。首先，当芯片制造商使用该工具来提高芯片成品率时，该工具可以带来显著的社会效益和环境效益。其次，为该产品开发的技术可能更广泛地适用于其他计算领域，如大型服务器系统、移动设备和消费电器；特别是在未来，细粒度的能源核算将成为所有计算设备不可或缺的一部分。除了对技术的影响外，该项目还将通过为参与该项目的学生提供教育福利来促进劳动力发展，这些学生将以类似于小企业的结构进行组织，鼓励他们在短时间内找到创新的解决方案，并将每周与业务导师会面。

This PFI: AIR Technology Translation project focuses on translating a novel integrated circuit (IC) power estimation simulation technology into an integrated simulation tool that is superior to the current state-of-the-art. The Scalable Accelerated Power Simulation Program with Integrated Circuit Emphasis (SAP-SPICE) will drastically improve the way IC manufacturers design, fabricate, and test modern chips by significantly increasing the speed and the accuracy of the power estimation simulations. This project addresses the following technology gap as it translates from research discovery toward commercial application. While the gold-standard simulation tool HSPICE produces the most accurate result, it comes at a high computational cost. On the other hand, accelerated simulators are faster because they depend on larger component models but they produce less accurate average results. The SAP-SPICE project uses a different approach with logic design, a logic simulator, and a portion of power data to extract a more accurate result while maintaining fast computational speed. The goal of this project is to produce a preproduction quality prototype that will integrate with existing industry standard tools to demonstrate the advantages of the technology over other tools to potential customers. The SAP-SPICE will have broader impacts on several fronts. First, the tool can lead to significant societal and environmental benefit when it is used by the chip manufacturers to improve the chip yield. Second, the techniques developed for the product have potential broader applicability to other areas of computing such as large server systems, mobile devices, and consumer appliances; especially in a future where a fine-grained energy accounting will be an integral part of all computing devices. Beyond its impact on technology, the project will also contribute to workforce development through educational benefits to the students participating in the project who will be organized in a structure similar to a small business, encouraged to find innovative solutions on a short timeline, and will meet weekly with the business mentor.