SHF: CSR: Small: CAD for THz lateral SiGe HBT on SOI to address Amdahl's Law

SHF：CSR：小型：SOI 上太赫兹横向 SiGe HBT 的 CAD，以解决阿姆达尔定律

• 批准号: 1618143
• 负责人: Mona Hella
• 金额: $ 45万
• 依托单位: Rensselaer Polytechnic Institute
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-07-15 至 2021-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1618143&HistoricalAwards=false
• 关键词: SHF; CSR; Small; CAD; THz

Modern microprocessors face a serious challenge. Their recent development cycles have not seen clock rate improvement, and their designs have relied upon so called multiple cores to produce enhanced performance. With a few exceptions, there has been little improvement in run time for many common applications, perhaps for as long as a decade testifying the failure of Moore's Law, on which the progress of computing industry is based. Due to its seemingly extreme speed, the present project explores a device, which may have the potential for breakthroughs in future multiple core computation at reasonable power levels. One of the important outreach goals of the research is to alert the integrated circuit design community (including students) of what is possible using these novel devices. The proposed research is focused on trying to make it easy for such designers to investigate this approach on their own, thus potentially resulting in extensive societal impact.The technical goal of this proposal is to investigate the use of lateral Silicon Germanium (SiGe) Heterojunction Bipolar Transistor (HBT) fabricated in a Silicon-on-Insulator (SOI) process. The impetus for this work comes from preliminary research demonstarting that these structures can drive capacitive loads (IC interconnect) at speeds exceeding 1 THz and with lower power than current and future FIN-FET CMOS devices. Specifically, the project will develop CAD tools for the lateral/SOI variety of high-speed SiGe heterojunction bipolar transistors (HBT).

现代微处理器面临着严峻的挑战。他们最近的开发周期没有看到时钟速率的改进，他们的设计依赖于所谓的多核来产生增强的性能。除了少数例外，许多常见应用程序的运行时几乎没有什么改进，可能长达十年的时间证明了摩尔定律的失败，而摩尔定律是计算行业进步的基础。由于其看似极端的速度，本项目探索的设备可能在合理的功率水平下在未来的多核计算中有突破的潜力。这项研究的重要目标之一是提醒集成电路设计界（包括学生）使用这些新设备的可能性。这项研究的重点是试图让这些设计师自己更容易地研究这种方法，从而可能产生广泛的社会影响。本提案的技术目标是研究在绝缘体上硅（SOI）工艺中制造的横向硅锗（SiGe）异质结双极晶体管（HBT）的使用。这项工作的动力来自于初步研究表明，这些结构可以以超过1太赫兹的速度驱动电容性负载（IC互连），并且比当前和未来的FIN-FET CMOS器件功耗更低。具体而言，该项目将开发用于横向/SOI各种高速SiGe异质结双极晶体管（HBT）的CAD工具。