An accurate and high-capacity electromagnetic solver for integrated circuit design

本研究的目的是开发一个高容量的电磁解算器的集成电路中的互连网络的设计，与AMD公司（AMD）合作。其目标是实现对集成电路的大部分的仿真，这在目前是不可行的。能够有效地执行大型仿真已经成为AMD等微电子公司的关键需求-这是任何公开的电磁求解器都无法满足的需求。该计划将开发一个创新的电磁求解器，研究新的算法来导入和网格化大型IC布局，新的方法来解决复杂几何形状的麦克斯韦方程，以及新的并行化策略。一个明确的重点将放在立即翻译拟议的发展，以生产，其目标是将它们集成到AMD的设计工作流程。这项研究计划的成果将是集成电路的电磁解算器，具有前所未有的准确性和可扩展性。新一代工程师也将接受培训，并配备电磁学，计算方法和高性能计算方面的先进技能。尽管在安大略和加拿大，微电子、信息技术、汽车、航空航天和量子技术等各个工业部门都在经历显著增长，但目前这类专业人士的需求非常高。

This research aims to develop a high-capacity electromagnetic solver for the design of interconnect networks in integrated circuits, in collaboration with Advanced Micro Devices (AMD). The goal is to enable the simulation of large portions of an integrated circuit, which is currently not feasible. Being able to perform large simulations efficiently has become a critical need of microelectronic companies like AMD - a need not met by any electromagnetic solver publicly available. The program will develop an innovative electromagnetic solver, investigating new algorithms to import and mesh large IC layouts, new methods to solve Maxwell equations in complex geometries, and new parallelization strategies. An explicit emphasis will be placed on immediately translating the proposed developments to production, with the goal of integrating them in AMD's design workflow. The outcome of this research program will be an electromagnetic solver for integrated circuits with unprecedented accuracy and scalability. A new generation of engineers will be also trained and equipped with advanced skills in electromagnetism, computational methods, and high-performance computing. There is currently a shortage of such professional figures, despite very high demand in Ontario and Canada from various industrial sectors undergoing significant growth: microelectronics, information technology, automotive, aerospace, and quantum technologies.