期刊文献+

阵列处理器动态可配置分布式存储访问结构设计

Design of dynamic configurable distributed storage access structure of array processor
下载PDF
导出
摘要 为了提高可重构阵列处理器的灵活性,打破传统片上互连的单一性。基于可重构阵列处理器分布式存储结构,提出一种动态可配置分布式存储访问结构。该结构可动态地将簇内4×4个PE独立地配置为本地访问模式、局部访问模式和全局访问模式,实现簇内4×4个PE对4×4个MB的并行访问。选用Xilinx公司的ZYNQ系列芯片XC7Z045 FFG900⁃2进行FPGA仿真。实验结果表明,该结构在无冲突访问模式下最高频率可达212.354 MHz,访问峰值带宽为7.125 GB/s。 In order to further improve the flexibility of reconfigurable array processors,the unicity of traditional on-chip interconnects is broken.A dynamic configurable distributed storage access structure is proposed on the basis of the distributed storage structure of reconfigurable array processor.The access structure can configure 4×4 PEs(processing elements)in the cluster dynamically and independently to local access mode,partly access mode,and global access mode according to the characteristics of data access.Moreover,the structure can achieve 4×4 PEs in the cluster to access 4×4 MBs(memory banks)in parallel.To verify the effectiveness of proposed structure,Xilinx’s ZYNQ series chip XC7 Z045 FFG900-2 is selected for FPGA simulation.The experiment results show that the maximum working frequency of the structure can reach 212.354 MHz and the access peak bandwidth can attain 7.125 GB/s in conflict-free access pattern.
作者 张园 刘有耀 山蕊 ZHANG Yuan;LIU Youyao;SHAN Rui(Xi’an University of Posts&Telecommunications,Xi’an 710121,China)
机构地区 西安邮电大学
出处 《现代电子技术》 2021年第12期11-15,共5页 Modern Electronics Technique
基金 国家自然科学基金资助项目(61802304) 国家自然科学基金资助项目(61772417) 国家自然科学基金资助项目(61834005) 国家自然科学基金资助项目(61602377) 国家自然科学基金资助项目(61634004) 国家自然科学基金资助项目(61874087) 陕西省国际科技合作计划项目(2018KW-006) 陕西省科技统筹创新工程项目(2016KTZDGY02-04-02) 陕西省重点研发计划(2017GY-060)。
关键词 阵列处理器 存储结构 分布式存储 访问模式 并行访问 实时动态配置 array processor storage structure distributed storage access pattern parallel access real-time dynamic configuration
  • 相关文献

参考文献2

二级参考文献37

  • 1Theodoridis G, Soudris D, and Vassiliadis S. A Survey of Coarse-grain Reconfigurable Architectures and Cad Tools[M] Netherlands: Springer, 2008: 89-149.
  • 2Mei B, Lambrechts A, Verkest D, et al.. Architecture exploration for a reconfigurable architecture template[J]. IEEE Design and Test of Computers, 2005, 22(2): 90-101.
  • 3Atak O and Atalar A. BilRC: an execution triggered coarse grained reconfigurable architecture[J]. IEEE Transactions on Very Large Scale Integration ( VLSI) Systems, 2012, 21(7): 1285-1298.
  • 4Yan M, Yang Z, Liu L, et al.. ProDFA: accelerating domain applications with a coarse-grained runtime reconfigurable architecture[C]. 18th IEEE International Conference on Parallel and Distributed Systems, Singapore, 2012: 834-839.
  • 5Ozaki N, Yasuda Y, Saito Y, et al.. Cool mega-arrays: ultralow-power reconfigurable accelerator chips[J]. IEEE Micro Magazine, 2011, 31(6): 6-18.
  • 6Suzuki T, Yamada H, Yamahishi T, et al.. High-throughput, low-power software-defined radio using reconfigurable processors[J]. IEEE Micro Magazine, 2011, 31(6): 19-28.
  • 7Singh H, Lee M-H, Lu G, et al.. MorphoSys: an integrated reconfigurable system for data-parallel and computation- intensive applications[J]. IEEE Transactions on Computers, 2000, 49(5): 465-481.
  • 8Ferreira R S, Cardoso J M P, Damiany A, et al.. Fast placement and routing by extending coarse-grained reconfigurable arrays with Omega networks[J]. Journal of System Architecture, 2011, 57(8): 761-777.
  • 9Cardoso J M P, Diniz P C, and Weinhardt M. Compiling for reconfigurable computing: a survey[J]. ACM Computer Survey, 2010, 42(4): 1-65.
  • 10Sutter B D, Coene P, Aa T V, et al.. Placement-and- routing- based register allocation for coarse-grained reconfigurable arrays[C]. Proceedings of the 2008 ACM SIGPLAN-SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems: Tucson, 2008: 151-160.

共引文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部