门级划分单片3D集成电路的时序驱动分析研磨机

Baoli Peng, Dongming Lv, Guojie Luo, M. Tahoori, Yuanqing Cheng
{"title":"门级划分单片3D集成电路的时序驱动分析研磨机","authors":"Baoli Peng, Dongming Lv, Guojie Luo, M. Tahoori, Yuanqing Cheng","doi":"10.1109/ICITES53477.2021.9637097","DOIUrl":null,"url":null,"abstract":"With the prevalence of data intensive applications like neuromorphic computing and big data processing, monolithic 3D (M3D) integration is proposed recently to provide ultrahigh memory bandwidth and can effectively improve the area, performance and power consumption of the chip. In this paper, we observe that it is essential to co-optimize the wirelength and critical path timing in placement for gate-level partitioned M3D ICs, which can take full advantage of dense MIVs and achieve better timing performance. Therefore, we propose a timing-driven analytical placement workflow for monolithic 3D ICs. Guided by timing analysis, the timing critical paths are selected and divided onto different tiers to exploit short vertical interconnects for timing optimization. Experimental results show that compared to TSV based 3D ICs, our timing-driven analytical placer for M3D ICs can reduce the chip footprint, the half-perimeter wirelength (HPWL) and the circuit delay by 70.2%, 55.6%, and 49.6% respectively.","PeriodicalId":370828,"journal":{"name":"2021 International Conference on Intelligent Technology and Embedded Systems (ICITES)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Timing-driven Analytical Placer for Gate-Level Partitioned Monolithic 3D ICs\",\"authors\":\"Baoli Peng, Dongming Lv, Guojie Luo, M. Tahoori, Yuanqing Cheng\",\"doi\":\"10.1109/ICITES53477.2021.9637097\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"With the prevalence of data intensive applications like neuromorphic computing and big data processing, monolithic 3D (M3D) integration is proposed recently to provide ultrahigh memory bandwidth and can effectively improve the area, performance and power consumption of the chip. In this paper, we observe that it is essential to co-optimize the wirelength and critical path timing in placement for gate-level partitioned M3D ICs, which can take full advantage of dense MIVs and achieve better timing performance. Therefore, we propose a timing-driven analytical placement workflow for monolithic 3D ICs. Guided by timing analysis, the timing critical paths are selected and divided onto different tiers to exploit short vertical interconnects for timing optimization. Experimental results show that compared to TSV based 3D ICs, our timing-driven analytical placer for M3D ICs can reduce the chip footprint, the half-perimeter wirelength (HPWL) and the circuit delay by 70.2%, 55.6%, and 49.6% respectively.\",\"PeriodicalId\":370828,\"journal\":{\"name\":\"2021 International Conference on Intelligent Technology and Embedded Systems (ICITES)\",\"volume\":\"23 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-10-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 International Conference on Intelligent Technology and Embedded Systems (ICITES)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICITES53477.2021.9637097\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Conference on Intelligent Technology and Embedded Systems (ICITES)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICITES53477.2021.9637097","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

随着神经形态计算和大数据处理等数据密集型应用的普及,最近提出了单片3D (M3D)集成技术,以提供超高的存储带宽,并能有效地提高芯片的面积、性能和功耗。在本文中,我们观察到,对于门级分区M3D集成电路来说,在放置时共同优化波长和关键路径时序是必不可少的,这样可以充分利用密集的miv并获得更好的时序性能。因此,我们提出了一个时序驱动的单片3D集成电路分析放置工作流程。在时序分析的指导下,选择并划分时序关键路径,利用短的垂直互联进行时序优化。实验结果表明,与基于TSV的3D集成电路相比,我们的时序驱动的M3D集成电路分析研磨机可将芯片占地面积、半周长(HPWL)和电路延迟分别降低70.2%、55.6%和49.6%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
A Timing-driven Analytical Placer for Gate-Level Partitioned Monolithic 3D ICs
With the prevalence of data intensive applications like neuromorphic computing and big data processing, monolithic 3D (M3D) integration is proposed recently to provide ultrahigh memory bandwidth and can effectively improve the area, performance and power consumption of the chip. In this paper, we observe that it is essential to co-optimize the wirelength and critical path timing in placement for gate-level partitioned M3D ICs, which can take full advantage of dense MIVs and achieve better timing performance. Therefore, we propose a timing-driven analytical placement workflow for monolithic 3D ICs. Guided by timing analysis, the timing critical paths are selected and divided onto different tiers to exploit short vertical interconnects for timing optimization. Experimental results show that compared to TSV based 3D ICs, our timing-driven analytical placer for M3D ICs can reduce the chip footprint, the half-perimeter wirelength (HPWL) and the circuit delay by 70.2%, 55.6%, and 49.6% respectively.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
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