An efficient Mesh-of-Tree based interconnect architecture for high performance 3D FPGAs

V. Pangracious, Z. Marrakchi, H. Mehrez
{"title":"An efficient Mesh-of-Tree based interconnect architecture for high performance 3D FPGAs","authors":"V. Pangracious, Z. Marrakchi, H. Mehrez","doi":"10.1109/ACQED.2015.7274000","DOIUrl":null,"url":null,"abstract":"In this paper we propose a three-dimensional (3D) interconnect network implementation based on a modified Mesh-of-Trees (MoT) topology for FPGA architecture design. To obtain the optimal MoT-based interconnect structure, the routing architecture of the 2D MoT-based FPGA is modified to include long routing segments that span multiple switch blocks in every row and column. By adjusting the percentage of long wire and span, a 2.5D or 3D high density MoT FPGAs can be designed. To design 3D multi-stacked MoT-based FPGAs, the 2D MoT FPGA is cut into two or more equal sections by adjusting the long wire span. The long wire segments are realized using 3D through silicon via (TSV). To design 2.5D interposer-based multi-FPGAs, we increase the number of cuts and apply appropriate optimization models to scale down the number of long wires and horizontal inter-FPGA interposer wires. Using our 2.5/3D design and simulation CAD flow, we demonstrate the speed and area of 3D MoT-based FPGA architecture is improved by 54% and 41% respectively in comparison to 3D Mesh-based FPGA.","PeriodicalId":376857,"journal":{"name":"2015 6th Asia Symposium on Quality Electronic Design (ASQED)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 6th Asia Symposium on Quality Electronic Design (ASQED)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ACQED.2015.7274000","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

In this paper we propose a three-dimensional (3D) interconnect network implementation based on a modified Mesh-of-Trees (MoT) topology for FPGA architecture design. To obtain the optimal MoT-based interconnect structure, the routing architecture of the 2D MoT-based FPGA is modified to include long routing segments that span multiple switch blocks in every row and column. By adjusting the percentage of long wire and span, a 2.5D or 3D high density MoT FPGAs can be designed. To design 3D multi-stacked MoT-based FPGAs, the 2D MoT FPGA is cut into two or more equal sections by adjusting the long wire span. The long wire segments are realized using 3D through silicon via (TSV). To design 2.5D interposer-based multi-FPGAs, we increase the number of cuts and apply appropriate optimization models to scale down the number of long wires and horizontal inter-FPGA interposer wires. Using our 2.5/3D design and simulation CAD flow, we demonstrate the speed and area of 3D MoT-based FPGA architecture is improved by 54% and 41% respectively in comparison to 3D Mesh-based FPGA.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
一种高效的基于Mesh-of-Tree的高性能3D fpga互连架构
在本文中,我们提出了一种基于改进的树网格(MoT)拓扑的三维(3D)互连网络实现,用于FPGA架构设计。为了获得最佳的mot互连结构,修改了二维mot FPGA的路由架构,使其包含跨越每行和每列多个交换块的长路由段。通过调整长导线和跨距的比例,可以设计出2.5D或3D高密度MoT fpga。为了设计3D多堆叠MoT FPGA,通过调整长线跨度将2D MoT FPGA切割成两个或多个相等的部分。长线段采用三维硅通孔(TSV)技术实现。为了设计基于2.5D中间层的多fpga,我们增加了切割数量,并应用适当的优化模型来减少长线和水平fpga间中间层线的数量。利用我们的2.5/3D设计和仿真CAD流程,我们证明了基于3D mot的FPGA架构的速度和面积分别比基于3D mesh的FPGA提高了54%和41%。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
A wide input voltage range start-up circuit for solar energy harvesting system Decision-based Biochips: A novel design for concurrent execution of networked bioassays integrated in scalable DMFBs Realization of non-linear i-v curve with low power dissipation using linear ion drift memristor model Clock gating assertion check: An approach towards achieving faster verification closure on clock gating functionality SynDFG: Synthetic dataflow graph generator for high-level synthesis
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1