SHF: Small: Research on New Challenges in EDA

微芯片是用于计算、通信、娱乐和其他消费电子产品的现代微电子系统的核心。为了设计和制造下一代复杂的微电子系统，需要在EDA（电子设计自动化）软件的设计上进行重大创新。该项目在微芯片和电路板层面解决了复杂微电子系统EDA中的四个新挑战：(1)超越芯片的EDA：今天的路由（布线）？由于没有现有的EDA软件可以解决这个问题，因此必须手动完成高密度复杂电路板。研究电路板布线以处理各种新技术问题。(2)光刻感知EDA：由于除了继续使用193nm光来打印（制造）32nm及以下尺寸的片上特征之外，没有其他可行的选择，因此精确打印变得极其困难。将开发EDA软件，以生产对光刻友好的设计，以成功地制造微芯片。(3) GPU EDA：图形处理单元（GPU）已成为最近流行的高性价比并行计算平台。如何利用GPU来加速关键的EDA任务是一个挑战，将被研究。(4)随机EDA：为了处理先进技术节点的过程变化，EDA软件将被设计为解决基本的图优化问题（例如，最短路径，最小生成树和网络流等），其中边权（成本）是随机变量。所提出的研究将推进EDA的知识。它还将为数学规划和组合优化等其他领域增加新的知识，因为最终的研究将需要解决大规模的优化问题。这个项目更广泛的影响包括技术进步和下一代工程师的教育。所提出的研究改进了微电子系统的设计和制造，对整个社会都有好处。新的研究成果将通过论文研究、课程设计、家庭作业和课堂教学传递给本科生和研究生。

Micro-chips are at the heart of modern microelectronic systems for computing, communication, entertainment, and other consumer electronics. In order to design and manufacture next generations of complex microelectronic systems, major innovations in the design of EDA (electronic design automation) software are needed. This project addresses four new challenges in EDA for complex microelectronic systems at both the micro-chip and the circuit board levels: (1) Beyond-die EDA: The routing (wiring) of today?s high-density complex circuit boards has to be done manually since no existing EDA software can solve the problem. Research will be carried out in circuit board routing to handle various new technology issues. (2) Litho-aware EDA: Since there is no alternative practical option but to continue using 193nm light to print (manufacture) on-chip features of size 32nm and below, accurate printing has become extremely difficult. EDA software will be developed to produce designs that are friendly to lithography for successful micro-chip manufacturing. (3) GPU EDA: Graphics processing unit (GPU) has become a popular cost-effective parallel computing platform recently. How to take advantage of GPU to accelerate critical EDA tasks is a challenge and will be studied. (4) Stochastic EDA: In order to handle process variations in advanced technology nodes, EDA software will be designed to solve fundamental graph optimization problems (e.g., shortest path, minimum spanning tree, and network flow etc.) where edge weights (costs) are random variables. The proposed research will advance knowledge in EDA. It will also add new knowledge to other fields such as mathematical programming and combinatorial optimization since ultimately the research will need to solve large scale optimization problems. The broader impacts of this project include technology advancement and the education of next generation of engineers. The proposed research improves the design and manufacturing of microelectronic systems which will benefit the society at large. New research results will be passed on to undergraduate and graduate students through dissertation research, course projects, homework, and classroom teaching.