期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于Amdahl定律扩展的多核处理器性能模型研究 被引量：2

Performance Model for Multicore Processor Based on Extended Amdahl's Law
下载PDF
导出
摘要 通过引入应用程序并行特征、通信开销、资源限制等因素,建立了基于Amdahl定律扩展的多核处理器性能模型.通过模型参数仿真,搜索面向特定应用的多核处理器设计空间,得出如下规律:增大计算核心规模可实现超线性加速比;结构应优先选择异构结构;设计多进程、大容量的共享通信区可降低核间通信开销;计算核心数目和规模由应用程序并行度和各并行部分比例及设计规模决定. Accounting uneven workload allocation,hardware bounds and communication latency, a performance analytic model is derived based on extended Amdahl＇ s law for multicore processors. The speedup model can be applied as a guide for hardware design as well as a performance prediction model in multicore processors domain. The design space for the specific multicore processors is explored. According to the results, some suggestions about how to design the specific multicore proces- sors, such as the architecture type （ homogenous or heterogeneous）, the number of cores integrated in the multicore system, the scale of cores to execute computing workload, and the communication area, are provided to the designers.
作者 冯晓 戴紫彬 蔡路亭 李伟
机构地区 解放军信息工程大学 复旦大学专用集成电路与系统国家重点实验室
出处 《电子学报》 EI CAS CSCD 北大核心 2017年第6期1424-1430,共7页 Acta Electronica Sinica
基金 国家自然科学基金(No.61404175)
关键词 多核处理器 性能模型 AMDAHL定律 并行计算 加速比 multicore processors Amdahl＇ s law performance model parallel computing speedup
分类号 TN492 [电子电信—微电子学与固体电子学]
  • 相关文献

参考文献3

二级参考文献52

  • 1Agarwal A. Tiled multicore processors: The four stages of reality [keynote][C] //Proc of the 40th Annual IEEE/ACM Int Symp on Microarchitecture (MICRO'07). Los Alamitos, CA: IEEE Computer Society, 2008.
  • 2Sankaralingam K, Nagarajan R, Gratz P, et al. The distributed microarehitecture of the TRIPS prototype processor [C] //Proc of the 39th Int Symp on Microarch tecture (MICRO06). Los Alamitos, CA:IEEE Computer Society, 2006:480-491.
  • 3Taylor M B, Lee W. Miller J, et al. Evaluation of the raw microprocessor: An exposed-wire-delay architecture for ILP and streams [C]//Proc of Int Symp on Computer Arehitecture(ISCA04). New York: ACM, 2004:2-13.
  • 4Butts M. Synchronization through communication in a massively parallel processor array [J]. IEEE Micro, 2007, 27(5) : 32-40.
  • 5Das R, Eachempati S, Mishra A K, et al. Design and evaluation of a hierarchical on chip interconnect for next generation CMPs [C] //Proe of IEEE 15th Int Symp on High Performance Computer Architecture (HPCA'09). Los Alamitos, CA: IEEE Computer Society, 2009: 175-186.
  • 6Balfour J, Dally W J. Design tradeoffs for tiled CMP on chip networks [C] //Proc of the 20th Annual Int Conf on Supercomputing(ICS'06). New York: ACM, 2006:187-198.
  • 7Bourduas S, Zilic Z. A hybrid ring/mesh interconnect of network-on-chip using hierarchical rings for global routing [C] //Proc of the 1st Int Symp on Networks-on Chips (NOCS'07). Los Alamitos, CA= IEEE Computer Society, 2007, 195-204.
  • 8Horowitz M, Dally W. How scaling will change processor architecture [C] //Proc of IEEE Int Solid-State Circuits Conf (ISSCC'04). Piscataway, NJ: IEEE, 2004:132-133.
  • 9Lu Z. Design and analysis of on-chip communication for network-on-chip platforms [D]. Stockholm: Department of Electronic Systems, Royal Institute of Technology (KTH), 2007.
  • 10Raghunathan V, Srivastava M B, Gupta R K. A survey of techniques for energy efficient on-chip communication [C] // Proc of the 40th Annual Design Automation Conf (DAC'03). New York: ACM, 2003 : 900-905.

共引文献20

同被引文献14

引证文献2

二级引证文献3

电子学报
2017年 第6期

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