Fast incremental link insertion in clock networks for skew variability reduction

A. Rajaram, D. Pan
{"title":"Fast incremental link insertion in clock networks for skew variability reduction","authors":"A. Rajaram, D. Pan","doi":"10.1109/ISQED.2006.66","DOIUrl":null,"url":null,"abstract":"With the advent of sub-100nm VLSI technologies, variation effects greatly increase the unwanted skew in clock distribution networks (CDNs), thereby reducing the performance of the chip. Recent works on link based non-tree CDN propose cross-link insertion in a given clock tree to reduce skew variation. However, the current methods suffer from the drawback that they are empirical in nature, requiring the user to experiment with different parameter values. Also, the methods of ignore the interaction between the different links while selecting the links for insertion. The method of (W.D. Lam et al., 2005) attempts to overcome this drawback using a statistical link insertion methodology. However, (W.D. Lam et al., 2005) is very slow even for relatively small circuits. In this paper, we propose a fast link insertion methodology which does not require selecting empirical parameters for link insertion. Our method also incrementally considers the effect of previously inserted links before choosing the next link. SPICE based Monte Carlo simulations show that our approach obtains comparable skew reduction to that of the existing approaches while drastically reducing the time taken to obtain a good link-based non-tree CDN","PeriodicalId":138839,"journal":{"name":"7th International Symposium on Quality Electronic Design (ISQED'06)","volume":"169 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2006-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"7th International Symposium on Quality Electronic Design (ISQED'06)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISQED.2006.66","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12

Abstract

With the advent of sub-100nm VLSI technologies, variation effects greatly increase the unwanted skew in clock distribution networks (CDNs), thereby reducing the performance of the chip. Recent works on link based non-tree CDN propose cross-link insertion in a given clock tree to reduce skew variation. However, the current methods suffer from the drawback that they are empirical in nature, requiring the user to experiment with different parameter values. Also, the methods of ignore the interaction between the different links while selecting the links for insertion. The method of (W.D. Lam et al., 2005) attempts to overcome this drawback using a statistical link insertion methodology. However, (W.D. Lam et al., 2005) is very slow even for relatively small circuits. In this paper, we propose a fast link insertion methodology which does not require selecting empirical parameters for link insertion. Our method also incrementally considers the effect of previously inserted links before choosing the next link. SPICE based Monte Carlo simulations show that our approach obtains comparable skew reduction to that of the existing approaches while drastically reducing the time taken to obtain a good link-based non-tree CDN
时钟网络中快速增量链路插入以减少倾斜可变性
随着sub-100nm VLSI技术的出现,变化效应大大增加了时钟分配网络(cdn)中不必要的倾斜,从而降低了芯片的性能。最近基于链路的非树CDN研究提出在给定时钟树中插入交叉链路以减少偏差。然而,目前的方法有一个缺点,即它们本质上是经验性的,需要用户用不同的参数值进行实验。此外,在选择要插入的链接时,方法忽略了不同链接之间的交互。(W.D. Lam et al., 2005)的方法试图使用统计链接插入方法来克服这一缺点。然而,(W.D. Lam et al., 2005)即使对于相对较小的电路也是非常慢的。在本文中,我们提出了一种不需要选择经验参数的快速链路插入方法。我们的方法还在选择下一个链接之前,增量地考虑先前插入的链接的影响。基于SPICE的蒙特卡罗模拟表明,我们的方法获得了与现有方法相当的倾斜减少,同时大大减少了获得良好的基于链接的非树CDN所需的时间
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信