吉林大学学报(工学版) ›› 2015, Vol. 45 ›› Issue (5): 1572-1580.doi: 10.13229/j.cnki.jdxbgxb201505028

• • 上一篇    下一篇

基于网络处理器的流媒体应用架构模型(VPL)

李明哲1, 2, 王劲林1, 陈晓1, 陈君1   

  1. 1.中国科学院声学研究所 国家网络新媒体工程技术研究中心,北京 100190;
    2.中国科学院大学 物理学院,北京 100190
  • 收稿日期:2014-03-10 出版日期:2015-09-01 发布日期:2015-09-01
  • 通讯作者: 陈君(1977-),女,副研究员.研究方向:流媒体技术.E-mail:chenj@dsp.ac.cn
  • 作者简介:李明哲(1988-）,男,博士研究生.研究方向:网络新媒体技术.E-mail:limz@dsp.ac.cn
  • 基金资助:
    国家科技支撑计划项目(2012BAH73F02); 中国科学院重点部署项目(KGZD-EW-103-4)

Architecture model of streaming media applications on network processors(VPL)

LI Ming-zhe1, 2, WANG Jin-lin1, CHEN Xiao1, CHEN Jun1   

  1. 1.National Network New Media Engineering Research Center, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China;
    2.School of Physics, University of Chinese Academy of Sciences, Beijing 100190, China
  • Received:2014-03-10 Online:2015-09-01 Published:2015-09-01

RICH HTML

0   

PDF (PC)

289

摘要： 研究了多核网络处理器上流媒体应用的软件架构,在混合流水线(HPL)模型的基础了提出了虚拟流水线(VPL)模型,该模型能够克服HPL负载难以均匀分配的问题。VPL模型将功能逻辑设计和资源分配相分离,增强了应用开发阶段的灵活性。为VPL提出了AF和NAF两种数据流映射策略,并对两种策略的性能进行了理论分析和实验测试, 探讨了二者之间如何选取。作为VPL的应用实例,基于VPL模型在Cavium OCTEON处理器上设计了流媒体网关应用HiliMG。对HiliMG的实验结果证明了VPL相对于HPL模型的性能优势,VPL使得吞吐率提升13%以上。

关键词: 计算机软件, 流媒体, 多核架构, 网络处理器, 软件流水线

Abstract: Designing of high-performance software architectures for streaming media applications on network processors is studied. The VPL (Virtual Pipeline) model is presented as an enhancement to the commonly used the HPL (Hybrid Pipeline) multi-core architecture model. VPL overcomes the load imbalance problem of HPL. Besides, VPL separates application logic design from resource mapping which improves flexibility of software development. VPL supports a variety of data flow mapping strategies. Two of such strategies are proposed. Performance of these strategies was analyzed theoretically and measured on real hardware to explore when to use each. As an example application of VPL, HiliMG, a streaming media gateway application, was designed and implemented on a Cavium OCTEON network processor. Experimental results on HiliMG further confirmed that VPL performs better than HPL in term of data throughput by at least 13%.

Key words: computer software, streaming media, multicore architecture, network processor, software pipeline

中图分类号: 

  • TP316
[1] 宋毅. 数据复制转发平滑引擎的多核网络软件框架技术研究[D]. 北京:中国科学院大学物理学院, 2013. Song Yi. The research of multicore network software framework of data copying relaying and smoothing engine[D]. Beijing: School of Physics, University of Chinese Academy of Sciences, 2013.
[2] Bae K, Ok S, Son H, et al. An Efficient Interworking Architecture of a Network Processor for Layer 7 Packet Processing[M].Communication and Networking: Springer, 2012:136-146.
[3] Meng J, Chen X, Chen Z, et al. Towards High-performance Ipsec on Cavium Octeon Platform[M]. Berlin:Trusted Systems, Spriger Berlin Heidelberg,2011:37-46.
[4] Li M, Zhang W, Chen X, et al. Performance evaluation of an ip-san initiator based on multi-core network processors[C]∥International Conference on Computer Technology and Development, 3rd (ICCTD 2011),ASME Press, 2011:1-5.
[5] Lee K. OpenNP : a generic programming model for network processors[D].Lancashire:Lancaster University, 2006.
[6] Plishker W, Keutzer K. NP-Click: a productive software development approach for network processors[J]. IEEEE Micro, 2004, 24(5):45-54.
[7] 闫守孟. 面向网络处理器的软件平台关键技术研究[D]. 西安:西北工业大学计算机学院, 2005. Yan Shou-meng. Key technologies study on network processors' software platform[D]. Xi'an:School of Computer Science,Northwestern Polytechnical University, 2005.
[8] 贺鹏程. 面向流的多核分组处理与传输技术研究[D]. 北京:中国科学院研究生院, 2010. He Peng-cheng. Studying flow based packet scheduling and transimission on multi-core processor[D]. Beijing: Graduate University of Chinese Academy of Sciences, 2010.
[9] Guo D, Liao G, Bhuyan L, et al. A scalable multithreaded L7-filter design for multi-core servers[C]∥Proceedings of the 4th ACM/IEEE Symposium on Architectures for Networking and Communications Systems. New York, USA: ACM Press, 2008:60.
[10] Ahuja V, Farrens M, Ghosal D. Cache-aware affinitization on commodity multicores for high-speed network flows[C]∥Proceedings of the Eighth ACM/IEEE Symposium on Architectures for Networking and Communications Systems, ACM, 2012:39-48.
[11] Cho J Y, Jin H W, Lee M, et al. On the core affinity and file upload performance of hadoop[C]∥Proceedings of the 2013 International Workshop on Data-Intensive Scalable Computing Systems.New York, USA: ACM Press, 2013:25-30.
[12] Gaud F, Geneves S, Lachaize R, et al. Efficient workstealing for multicore event-driven systems[C]∥2010 IEEE 30th International Conference on Distributed Computing Systems,Genova,2010: 516-525.
[13] Haller P, Odersky M. Scala actors: unifying thread-based and event-based programming[J]. Theoretical Computer Science, 2009, 410(2/3): 202-220.
[14] Cavium networks[EB/OL]. [2014-03-07].http://www.cavium.com/
[15] Breznick A. A switch in time: the role of switched digital video in easing the looming bandwidth crisis in cable[J]. Communications Magazine, IEEE, 2008, 46(7):96-102.
[16] HiliMG[EB/OL]. [2014-03-07].http://www.hiliway.com/zh/products/hilimg.html
[17] Li J, Chen J, Li M, et al. A multi-core architecture for video streaming[J]. Applied Mechanics and Materials, 2013, 411:960-965.
[18] 郭秀岩. 面向多核的多层次实时网络数据流调度技术研究[D]. 合肥:中国科学技术大学信息科技学院, 2011. Guo Xiu-yan. The research of multi-level real-time network stream scheduling on multicore network processor[D]. Hefei: School of Information Science and Technology,University of Science and Technology of China, 2011.
[19] Zhou Y. Hardware acceleration for power efficient deep packet inspection[D].Dublin:Dublin City University, 2012.
[1] 侯永宏, 王利伟, 邢家明. 基于HTTP的动态自适应流媒体传输算法[J]. 吉林大学学报(工学版), 2018, 48(4): 1244-1253.
[2] 马健, 樊建平, 刘峰, 李红辉. 面向对象软件系统演化模型[J]. 吉林大学学报(工学版), 2018, 48(2): 545-550.
[3] 罗养霞, 郭晔. 基于数据依赖特征的软件识别[J]. 吉林大学学报(工学版), 2017, 47(6): 1894-1902.
[4] 应欢, 王东辉, 武成岗, 王喆, 唐博文, 李建军. 适用于商用系统环境的低开销确定性重放技术[J]. 吉林大学学报(工学版), 2017, 47(1): 208-217.
[5] 李勇, 黄志球, 王勇, 房丙午. 基于多源数据的跨项目软件缺陷预测[J]. 吉林大学学报(工学版), 2016, 46(6): 2034-2041.
[6] 王念滨, 祝官文, 周连科, 王红卫. 支持高效路径查询的数据空间索引方法[J]. 吉林大学学报(工学版), 2016, 46(3): 911-916.
[7] 特日跟, 江晟, 李雄飞, 李军. 基于整数数据的文档压缩编码方案[J]. 吉林大学学报(工学版), 2016, 46(1): 228-234.
[8] 康辉, 王家琦, 梅芳. 基于Pi演算的并行编程语言[J]. 吉林大学学报(工学版), 2016, 46(1): 235-241.
[9] 陈鹏飞, 田地, 杨光. 基于MVC架构的LIBS软件设计与实现[J]. 吉林大学学报(工学版), 2016, 46(1): 242-245.
[10] 刘衍珩, 李松江, 王爱民. P2P流媒体中动态分级传输模型及传输算法[J]. 吉林大学学报(工学版), 2016, 46(1): 259-264.
[11] 刘磊, 王燕燕, 申春, 李玉祥, 刘雷. Bellman-Ford算法性能可移植的GPU并行优化[J]. 吉林大学学报(工学版), 2015, 45(5): 1559-1564.
[12] 冯晓宁, 王卓, 张旭. 基于L-π演算的WSN路由协议形式化方法[J]. 吉林大学学报(工学版), 2015, 45(5): 1565-1571.
[13] 王克朝, 王甜甜, 苏小红, 马培军. 基于频繁闭合序列模式挖掘的学生程序雷同检测[J]. 吉林大学学报(工学版), 2015, 45(4): 1260-1265.
[14] 黄宏涛，王静，叶海智，黄少滨. 基于惰性切片的线性时态逻辑性质验证[J]. 吉林大学学报(工学版), 2015, 45(1): 245-251.
[15] 范大娟1, 2, 黄志球1, 肖芳雄1, 祝义1, 王进1. 面向多服务交互的相容性分析与适配器生成[J]. 吉林大学学报(工学版), 2014, 44(4): 1094-1103.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发