期刊文献+

实时信息处理系统多线程框架的高效设计方法 被引量:7

An efficient multi-thread framework designing method for real-time information processing systems
下载PDF
导出
摘要 针对目前的多线程编程模型不能快速有效地解决实时信息处理系统(RIPS)设计的问题,通过对以流水线方式工作的多线程系统进行建模,分析了两种适用于RIPS的多线程模型的时间开销与线程数量、缓冲区数量之间的关系,并基于这两种多线程模型,提出了一种高效的RIPS多线程设计方法。该方法无需过多考虑各线程的设计细节,从而能有效提高多线程RIPS的开发效率,同时保证系统的高效与稳定。在x86通用PC机平台上对该方法进行的仿真验证的结果表明,该方法能够保证RIPS的可靠性和稳定性。与基于MPI搭建的系统的对比测试结果表明,基于该方法搭建的系统具有更高的效率和鲁棒性。 The study was conducted as follows to solve the problem that existing multi-threading programming models can- not be used to effectively design a real-time information processing system (RIPS) : A pipeline-wise multi-thread architecture was modeled, and the relationship between the overhead time and the number of threads or buffers in two multi-thread models suitable for RIPS was analyzed, and then, an efficient multi-thread programming method for design of RIPS was proposed based on these two models. The results of the experiment performed on the x86 based PC platform show that the proposed method can guarantee RIPS' efficient and smooth running. And the com- parative test shows that the RIPS designed based on the proposed method has the higher efficiency and robustness than that designed based on the programming framework of message passing interface (MPI). The method need not to consider the design details of each thread, so it can improve the efficiency of RIPS design and bring the system a good stability.
出处 《高技术通讯》 CAS CSCD 北大核心 2013年第1期42-48,共7页 Chinese High Technology Letters
基金 863计划(2009AA8012320B)资助项目.
关键词 实时信息处理系统(RIPS) 多线程 线程同步 流水线 real-time information processing system (RIPS), multi-thread, thread synchronization, pipeline
  • 相关文献

参考文献13

  • 1An K H,Chung M. Cognitive face analysis system for future interactive TV[J].IEEE Transactions on Consumer Electronics,2009,(04):2271-2279.
  • 2Garcia-Quinchia A,Guo Y,Martin E. A systemon-chip (SOC) platform to integrated inertial navigation systems&GPS[A].Seoul,Korea,2009.603-608.
  • 3Yu Y P. Object oriented teleconsultations in global PACS using multi-thread Java[A].Wailea,USA,1997.166-175.
  • 4Ma J;Ji Z;Cui M.Design and implementation of the web-based real-time remote expert identification system:used for biological quarantine[A]湖北武汉,20101-4.
  • 5Takeda R;Nakadai K;Takahashi T.Speedup and performance improvement of ica-based robot audition by parallel and resampling based block-wise processing[A]台湾台北,20101949-1956.
  • 6Traff J L,Gropp W D,Thakur R. Self-consistent MPI performance guidelines[J].IEEE Transactions on Parallel and Distributed Systems,2010,(05):698-709.doi:10.1109/TPDS.2009.120.
  • 7Ayguade E,Copty N,Duran A. The design of OpenMP tasks[J].IEEE Transactions on Parallel and Distributed Systems,2009,(03):404-418.
  • 8Gropp W,Lusk E. Goals guiding design:PVM and MPI[A].Chicago,USA,2002.257-265.
  • 9Stankovic N,Zhang K. A distributed parallel programming framework[J].IEEE Transactions on Software Engineering,2002,(05):478-493.doi:10.1109/TSE.2002.1000451.
  • 10Choi G S,Das C R. A superscalar software architecture model for multi-core processors (MCPS)[J].System Software,2010.1823-1837.

二级参考文献10

  • 1韩笑,李斌,田志宏.大规模网络环境下高可扩展的信息监测系统[J].计算机工程与应用,2005,41(25):152-154. 被引量:1
  • 2张志斌,郭莉,方滨兴,陈小军.一种基于自动机分解的网络协议并行处理策略[J].计算机学报,2006,29(8):1496-1504. 被引量:4
  • 3Masmano M, Ripoll I, Crespo A. A comparison of memory allocators for realtime applications. In: Proceedings of the 4th international workshop on Java technologies for real-time and embedded systems, Paris, France, 2006, 68-76
  • 4Gloger W. Dynamic memory allocator implementations in Linux system libraries, http://www. dent. med. uni-muenchen. de/wmglo/malloc-shdes.html, 1998
  • 5Chuck L, David B. Malloc( ) Performance in a Multithreaded Linux Environment: [CITI Technical Report 00-5]. Ann Arbor: Center for Information Technology Integrity of University of Michigan, 2000
  • 6Larson P A, Krishnan M. Memory allocation for long-running server applications. In: Proceedings of the International Symposium on Memory Management, Vancouver, B.C., Canada, 1998, 176-185
  • 7Berger E D, MCKINLEY K S, Blumofe R D, et al. Hoard: A Scalable Memory Allocator for Muhithreaded Applications. In: Proceeding of the International Conference on Arehitectural Support for Programming Languages and Operating Systems (ASPLOS-IX). SIGPLAN Notices, Cambridge, Massachussetts, 2000, 117-128
  • 8Sanjay G, Paul M. TCMalloc: Thread-Caching Malloc. http://goog-perftools. sourceforge. net/doc/tcmalloc.html, 2005
  • 9Lea D. A Memory Allocator. http://gee. cs. oswego. edu/dl/ html/malioc.html, 2003
  • 10罗浩,云晓春,方滨兴.多线程TCP/IP协议还原技术的研究[J].高技术通讯,2003,13(11):15-19. 被引量:6

共引文献1

同被引文献79

  • 1吴君青,陈卫卫,胡谷雨.物理网络拓扑发现算法的研究与改进[J].北京邮电大学学报,2003,26(z2):139-144. 被引量:5
  • 2赵滨杰,朱明,邵中平.速度滑冰运动员训练监控管理系统的开发与应用[J].冰雪运动,2013,35(6):1-4. 被引量:4
  • 3朱有产,李春祥.一种跨VLAN的网络拓扑发现算法[J].计算机工程,2005,31(3):134-136. 被引量:8
  • 4黄乐.一个基于多线程的.NET开发应用实例研究[J].计算机与现代化,2006(7):59-61. 被引量:4
  • 5SHI C B, WANG G J, PEI X K, et al. An interleaving updating frame- work of disparity and confidence map for stereo matching [ J ]. IEICE Trans. Infomlation and Systems,2012(5) : 1552-1555.
  • 6FILHO G G. An optimal time-space algorithm for dense stereo matching [J]. Journal of Real-Time Image Processing,2012,7(2) :69-86.
  • 7LI J. An image feature point matching algorithm based on fixed scale fea- ture transformation[ J ]. Optik-lntemational Journal for Light and Elec-tron Optics,2013,124(13) :1620-1623.
  • 8ZHOU Y H, CHEN Y Q. Feature matching for automated and reliable initialization in three-dimensional digital image correlation [ J]. Optics and Lasers in Engineering,2013,51 (3) :213-223.
  • 9XIANG X Q,ZHANG M M,LI G X, et al. Real-time stereo matching based on fast belief propagation [ J ]. Machine Vision and Applications, 2012, 23(6) :1219-1227.
  • 10PEREZ J M,SANCHEZ P, MARTINEZ M,et al. Real-time stereo matc- hing using memory-efficient Belief Propagation for high-definition 3D telepresence systems [ J ]. Pattern Recognition Letters, 2011,32 ( 16 ) : 2250-2253.

引证文献7

二级引证文献12

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