当前位置:首页 |  学术交流学术交流

陈耿杰博士报告

信息来源: 暂无 发布日期: 2019-09-01 浏览次数:

主讲人:陈耿杰博士  

报告时间:2019年9月4号,9:30-11:00

报告地点:数计学院4号楼229

报告题目1:VLSI  Routing: Seeing Nano Tree in Giga Forest

报告摘要:We  are using nanometer-sized transistors and gigahertz clock frequency in very  large scale integration (VLSI). Under such extreme conditions, timing, power,  manufacturability and reliability are all crucial issues in VLSI design. The  tree structure is the major topology used in VLSI routing. Optimizing the tree  and the forest is essential for a successful design flow. However, the problems  are in general challenging. First, even for many single-net routing problems,  finding an optimal tree from a huge candidate forest is already NP-hard. Second,  for multiple-net routing, a large number of trees need to be built on the chip  by sharing resources and need to be well coordinated for avoiding conflicts.  Third, in order to achieve a full-flow success, it is also necessary to foresee  routing trees and to consider routability in early stages (e.g., placement).  This talk will introduce the efficient and effective algorithms that we devised  for tackling the three levels of challenges with not only practical  considerations in VLSI design but also mathematical rigorousness and guarantee.  

个人简介:Gengjie  Chen is currently a Principle Software Engineer with Giga Design Automation,  Shenzhen. He received the Ph.D. degree from the Department of Computer Science  and Engineering, The Chinese University of Hong Kong, Hong Kong, in 2019, and  received the B.Sc. degree from the Department of Electronic and Communication  Engineering, Sun Yat-sen University, Guangzhou, China, in 2015. His research  interests include combinatorial optimization, numerical optimization and  electronic design automation. He published 14 papers in TCAD, DAC, ICCAD, DATE  and ASPDAC during his Ph.D. study. He also received the first place at ACM  Student Research Competition in 2019, the first place at ACM SIGDA Student  Research Competition in 2018, the best paper award at ICCAD 2017, the Hong Kong  Ph.D. Fellowship from 2015 to 2019, and seven ICCAD/ISPD contest awards  including four championships. He has interned in Cadence, Austin, TX, in 2017,  and Synopsys, Mountain View, CA, in 2016.