FPGA based path solvers for DFGs in high level synthesis

D. Yagain, A. Vijayakrishna, P. Nikhil, A. Adarsh, S. Karthikeyan
{"title":"FPGA based path solvers for DFGs in high level synthesis","authors":"D. Yagain, A. Vijayakrishna, P. Nikhil, A. Adarsh, S. Karthikeyan","doi":"10.1109/ICTEA.2012.6462882","DOIUrl":null,"url":null,"abstract":"Retiming is a transformation which can be applied to Digital Signal Processing Blocks that can increase the clock frequency. Folding in retiming can also reduce the resource utilization and power consumption. This transformation requires computation of critical path and shortest path at various stages. In this particular work, a FPGA based path finder is designed to compute critical path and shortest path in the Data Flow Graphs (DFGs). Since this path computation is performed using FPGA based IC, the speed of retiming transformation increases. This also reduces the resource utilization of the general purpose machine in which retiming transformation is usually performed. Critical path in sequential circuit is defined as the longest path between any two storage components. This determines the minimum feasible clock period for any sequential circuit. We need to compute the critical path before we apply retiming transformation to any digital Signal Processing block. Similarly shortest path computation is required in retiming while solving the system inequalities in the constraint graph. In this work, shortest path computation is performed using Floyd-Warshall algorithm. Since FPGA based hardware for path solvers performs much faster when compared to general purpose processor [where actual retiming is done], the speed with which the retiming transformation is performed increases. Xilinx ISE design suit is used with device as SPARTEN3E XC3S250E for the work presented.","PeriodicalId":245530,"journal":{"name":"2012 2nd International Conference on Advances in Computational Tools for Engineering Applications (ACTEA)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 2nd International Conference on Advances in Computational Tools for Engineering Applications (ACTEA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICTEA.2012.6462882","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

Abstract

Retiming is a transformation which can be applied to Digital Signal Processing Blocks that can increase the clock frequency. Folding in retiming can also reduce the resource utilization and power consumption. This transformation requires computation of critical path and shortest path at various stages. In this particular work, a FPGA based path finder is designed to compute critical path and shortest path in the Data Flow Graphs (DFGs). Since this path computation is performed using FPGA based IC, the speed of retiming transformation increases. This also reduces the resource utilization of the general purpose machine in which retiming transformation is usually performed. Critical path in sequential circuit is defined as the longest path between any two storage components. This determines the minimum feasible clock period for any sequential circuit. We need to compute the critical path before we apply retiming transformation to any digital Signal Processing block. Similarly shortest path computation is required in retiming while solving the system inequalities in the constraint graph. In this work, shortest path computation is performed using Floyd-Warshall algorithm. Since FPGA based hardware for path solvers performs much faster when compared to general purpose processor [where actual retiming is done], the speed with which the retiming transformation is performed increases. Xilinx ISE design suit is used with device as SPARTEN3E XC3S250E for the work presented.
基于FPGA的DFGs高级综合路径求解器
重定时是一种变换,可以应用于数字信号处理块,可以提高时钟频率。折叠重新计时也可以减少资源利用率和功耗。这种变换需要计算各个阶段的关键路径和最短路径。在这个特殊的工作中,设计了一个基于FPGA的寻径器来计算数据流图(DFGs)中的关键路径和最短路径。由于该路径计算是使用基于FPGA的集成电路进行的,因此提高了重定时变换的速度。这也减少了通常执行重定时转换的通用机器的资源利用率。顺序电路中的关键路径是指任意两个存储元件之间的最长路径。这决定了任何顺序电路的最小可行时钟周期。在对任何数字信号处理块应用重定时变换之前,我们需要计算关键路径。同样,在求解约束图中的系统不等式时,也需要进行最短路径计算。在这项工作中,最短路径计算使用Floyd-Warshall算法进行。由于基于FPGA的路径求解器硬件比通用处理器执行得快得多,因此执行重定时转换的速度增加了。赛灵思ISE设计套装与器件SPARTEN3E XC3S250E配合使用。
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:604180095
Book学术官方微信