全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...
电子学报  2015 

基于种类-位置模型的可重构资源指派方法

DOI: 10.3969/j.issn.0372-2112.2015.02.015, PP. 299-304

Keywords: 可重构计算,资源建模,资源指派,定制指令

Full-Text   Cite this paper   Add to My Lib

Abstract:

在基于指令集动态可扩展技术的可重构指令集处理器研究中,如何有效使用系统的可重构资源,将很大程度上影响扩展得到的定制指令的功能实现,进而影响系统性能的优化效果.本文针对可重构资源的利用问题,首先设计了一种可重构资源模型,该模型弱化了可重构资源的功能和数量属性,主要提供其种类和位置属性,并能够以此计算资源使用的时间属性.基于此模型,本文将图论中的图着色问题进行扩展,引入多遍着色的思想,提出了一种针对粗粒度可重构资源的资源指派算法,该算法将可重构资源的指派等价为一个图多遍着色问题,通过模型提供的属性参数和限制条件完成指派过程.实验结果验证了算法的有效性,并揭示了资源使用中的规律性,对提高资源利用率和系统性能具有一定的指导意义.

References

[1]  Jain M K,Balakrishnan M,Kumar A.ASIP design methodologies:survey and issues[A].Proceeding of 14th International Conference on VLSI Design[C].Los Alamitos,CA:IEEE Computer Society,2001.76-81.
[2]  Galuzzi C,Bertels K.The instruction-set extension problem:a survey[J].ACM Transactions on Reconfigurable Technology and Systems,2011,4(2):1-28.
[3]  Barat F,Lauwereins R.Reconfigurable instruction set processors:a survey[A].Proceedings of 11th International Workshop on Rapid System Prototyping[C].Los Alamitos,CA:IEEE Computer Society,2000.168-173.
[4]  Vasicek Z,Sekanina L.An evolvable hardware system in Xilinx Virtex II Pro FPGA[J].International Journal of Innovative Computing and Applications,2007,1(1):63-73.
[5]  Lewis D,et al.The Stratix II logic and routing architecture[A].Proceedings of 13th International Symposium on Field-Programmable Gate Arrays[C].New York:ACM,2005.14-20.
[6]  Li X,Yang H,Zhong H.Use of VPR in design of FPGA architecture[A].Proceedings of 8th International Conference on Solid-State and Integrated Circuit Technology[C].Shanghai:IEEE Press,2006.1880-1882.
[7]  Lagadec L,Pottier B.Object-oriented meta tools for reconfigurable architectures[A].Proceedings of 2000 International Conference on Modeling,Signal Processing and Control[C].Bellingham WA:SPIE,2000.69-79.
[8]  Bossuet L,Gogniat G,et al.A modeling method for reconfigurable architectures[A].Proceedings of 2nd International Workshop on System-on-Chip for Real-Time Applications[C].Alberta,Canada:IEEE Computer Society,2002.170-179.
[9]  Li Z,et al.A modeling and mapping method for coarse/fine mixed-grained reconfigurable architecture[A].Proceedings of 11th International Conference on Solid-State and Integrated Circuit Technology[C].Xi''an:IEEE Press,2012.1-4.
[10]  Ito K.A scheduling and allocation method to reduce data transfer time by dynamic reconfiguration[A].Proceedings of 2000 Asia and South Pacific Design Automation Conference[C].New York:ACM,2000.323-328.
[11]  Eguro K.Resource allocation for coarse-grain FPGA development[J].IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems,2006,24(10):1572-1581.
[12]  Plaxton C G,Yu S,et al.Reconfigurable resource scheduling[A].Proceedings of 18th Annual ACM Symposium on Parallelism in Algorithms and Architectures[C].New York:ACM,2006.93-102.
[13]  Cavrilovska A,Kumar S,et al.High-performance hypervisor architectures:virtualization in HPC systems[A].Proceedings of 1st Workshop on System-Level Virtualization for High Performance Computing[C].New York:ACM,2007.1-8.
[14]  Shafigue M,Bauer L,et al.Minority-game-based resource allocation for run-time reconfigurable multi-core processors[A].Proceedings of 2011 Design,Automation & Test in Europe[C].New York:ACM,2011.1-6.
[15]  Li J,Das A,Kumar A.A design flow for partially reconfigurable heterogeneous multi-processor platforms[A].Proceedings of 23rd IEEE International Symposium on Rapid System Prototyping[C].Tampere,Finland:IEEE Reliability Society,2012.170-176.
[16]  齐骥,李曦,等.基于硬件任务顶点的可重构系统资源管理算法[J].电子学报,2006,34(11):2094-2098. Qi Ji,Li Xi,et al.Algorithms ofresource management for reconfigurable systems based on hardware task vertexes[J].Acta Electronica Sinica,34(11):2094-2098.(in Chinese)
[17]  周学海,等.基于差分进化和贪心策略的自定义指令选择算法研究[J].电子学报,2009,37(2):372-376. Zhou Xue-hai,et al.Study on differential evolution and greedy strategy based custom instruction selection algorithms[J].Acta Electronica Sinica,2009,37(2):372-376.(in Chinese)

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133