基于Amdahl定律的多核密码处理器性能模型研究

冯晓; 戴紫彬; 李伟; 蔡路亭

doi:10.11999/JEIT150474

基于Amdahl定律的多核密码处理器性能模型研究

doi: 10.11999/JEIT150474

1.
(解放军信息工程大学郑州 450000) ②(复旦大学专用集成电路与系统国家重点实验室上海 201203)

基金项目:

国家自然科学基金项目(61404175)

计量
- 文章访问数: 1564
- HTML全文浏览量: 145
- PDF下载量: 445
- 被引次数: 0
出版历程
- 收稿日期: 2015-04-27
- 修回日期: 2015-12-25
- 刊出日期: 2016-04-19

Performance Model of Multicore Crypto Processor Based on Amdahls Law

1.
(PLA Information Engineering University, Zhengzhou 450000, China)

Funds:

National Natural Science Foundation of China (61404175)

摘要

摘要: 该文构建面向密码应用的多核处理器性能模型，对多核密码处理器设计提供理论支持和有效建议。通过引入密码并行处理特征、数据传输时间、同步时间等因素，建立基于Amdahl定律扩展的多核密码处理器性能模型，基于提出的性能模型，对多核密码处理器设计空间进行搜索。模拟分析表明，影响多核密码处理器性能的关键因素是密码应用的可开发并行度、并行部分所占比例以及运算过程的通信次数。
- 密码处理器 /
- 多核处理器 /
- Amdahl定律 /
- 性能模型 /
- 通信/计算比
Abstract: This paper builds a performance model of multicore processor, which applies to the crypto algorithms, and some advisable suggestions and academic supports are given for its design. By introducing parallelism degree, transformation overhead and synchronization time, performance model of multicore crypto processor is built based on the Amdahls law and its extension, and accordingly the design space of multicore crypto processors is searched. Simulation analysis shows that the key factors influencing the performance model of multicore crypto processor are the exploitable parallelism of crypto application, the proportion of parallel part and the communication times in the process of operation.
- Crypto processor /
- Multicore processor /
- Amdahls law /
- Performance model /
- Communication/computing ratio

HTML全文

参考文献(25)

BOSSUET L, GRAND M, GASPAR L, et al. Architectures of flexible symmetric key crypto engines-a survey: from hardware coprocessor to multi-crypto-processor system on chip[J]. ACM Computing Surveys (CSUR), 2013, 45(4): 1-32. doi: 10.1145/2501654.2501655.

冯晓静. 面向服务的异构多核片上系统的关键技术研究及实现[D]. [博士论文], 中国科学技术大学, 2013.

FENG Xiaojing. Study and implementation of service oriented heterogeneous multi processor system-on-chip[D]. [Ph.D. dissertation], University of Science and Technology of China, 2013.

蒋骁辰, 李国平, 王国中, 等. 基于AVS+实时编码的多核并行视频编码算法[J]. 电子与信息学报, 2014, 36(4): 810-816. doi: 10.3724/SP.J.1146.2013.00845.

JIANG Xiaochen, LI Guoping, WANG Guozhong, et al. Multi-core parallel video coding algorithm based on AVS+real-time encoding[J]. Journal of Electronics Information Technology, 2014, 36(4): 810-816. doi: 10.3724/ SP.J. 1146.2013.00845.

SHUKLA S K, MURTHY C N S, and Chande P K. A Survey of Approaches used in Parallel Architectures and Multi-core Processors, for Performance Improvement[M]. Switzerland, Springer International Publishing, 2015: 537-545.

SILBERSTEIN M. GPUs: High-performance accelerators for parallel applications: the multicore transformation (ubiquity symposium)[J]. Ubiquity, 2014, 2014: 1-13. doi: 10.1145/ 2618401.

AMDAHL G M. Validity of the single processor approach to achieving large scale computing capabilities[C]. Proceedings of Spring Joint Computer Conference, New York, 1967: 483-485.

刘斌, 赵银亮, 韩博, 等. 基于性能预测的推测多线程循环选择方法[J]. 电子与信息学报, 2014, 36(11): 2768-2774. doi: 10.3724/SP.J.1146.2013.01879.

LIU Bin, ZHAO Yinliang, HAO Bo, et al. A loop selection approach based on performance prediction of speculative multithreading[J]. Journal of Electronics Information Technology, 2014, 36 (11): 2768-2774. doi: 10.3724/SP.J.1146. 2013.01879.

EYERMAN S and EECKHOUT L. Modeling critical sections in Amdahl's law and its implications for multicore design[C]. ACM SIGARCH Computer Architecture News, New York, 2010: 362-370.

陈书明, 陈胜刚, 尹亚明. Amdahl 定律在层次化片上多核处理器中的扩展[J]. 计算机研究与发展, 2012, 49(1): 83-92.

CHEN Shuming, CHEN Shenggang, and YIN Yaming. Revisting Amdahls law in the hierarchical chip multicore processors[J]. Journal of Computer Research and Development, 2012, 49(1): 83-92.

TANG S, LEE B S, and HE B. Speedup for multi-Level parallel computing[C]. Parallel and Distributed Processing Symposium Workshops PhD Forum (IPDPSW), Shanghai, 2012: 537-546.

JUURLINK B H H and MEENDERINCK C H. Amdahl's law for predicting the future of multicores considered harmful[J]. ACM SIGARCH Computer Architecture News, 2012, 40(2): 1-9. doi: 10.1145/2234336.2234338.

KHANYILE N P, TAPAMO J R, and DUBE E. An analytic model for predicting the performance of distributed applications on multicore clusters[J]. IAENG International Journal of Computer Science, 2012, 39(3): 312-320.

CASSIDY A S and ANDEROU A G. Beyond Amdahl's law: an objective function that links multiprocessor performance gains to delay and energy[J]. IEEE Transactions on Computers, 2012, 61(8): 1110-1126. doi: 10.1109/TC.2011. 169.

YAVITS L, MORAD A, and GINOSAR R. The effect of communication and synchronization on Amdahls law in multicore systems[J]. Parallel Computing, 2014, 40(1): 1-16. doi: 10.1016/j.parco.2013.11.001.

CHE H and NGUYEN M. Amdahls law for multithreaded multicore processors[J]. Journal of Parallel and Distributed Computing, 2014, 74(10): 3056-3069. doi: 10.1016/j.jpdc. 2014.06.012.

AL-BABTAIN B M, AL-KANDERI F J, Al-Fahad M F, et al. A survey on Amdahl's law extension in multicore architectures[J]. International Journal of New Computer Architectures and their Applications (IJNCAA), 2013, 3(3): 30-46.

HILL M D and MARTY M R. Amdahl's Law in the Multicore Era[J]. Computer, 2008, 41(7): 33-38. doi: 10.1109/MC.2008. 209.

ASANOVI?C K, BODIK R, CATANZARO B C, et al. The landscape of parallel computing research: A view from Berkeley[R]. Technical Report of Electrical Engineering and Computer Sciences University of California at Berkeley, Berkeley: UC Berkeley, 2006: 8-11.

BUCHTY R, HEINTAE N, and OLIVA D. Cryptonite-A Programmable Crypto Processor Architecture for High- bandwidth Applications[M]. Berlin Heidelberg Springer, 2004: 184-198.

徐卫志, 宋风龙, 刘志勇, 等. 众核处理器片上同步机制和评估方法研究[J]. 计算机学报, 2010, 33(10): 1777-1787. doi: 10.3724/SP.J.1016. 2010.01777.

XU Weizhi, SONG Fenglong, LIU Zhiyong, et al. On synchronization and evaluation method of chipped many-core processor[J]. Chinese Journal of Computers, 2010, 33(10): 1777-1787. doi: 10.3724/SP.J.1016.2010.01777.

施引文献

资源附件(0)

访问统计