Advertisement

Replication-Based Partial Dynamic Scheduling on Heterogeneous Network Processors

  • Zhiyong Yu
  • Zhiyi Yang
  • Fan Zhang
  • Zhiwen Yu
  • Tuanqing Zhang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4847)

Abstract

It is a great challenge to map network processing tasks to processing resources of advanced network processors, which are heterogeneous and multi-threading multiprocessor System-on-Chip. This paper proposes a novel scheduling algorithm, called Replication-based Partial Dynamic Scheduling (RPDS). It aims to improve the NP performance by combining the strategies of partial dynamic mapping and task replication with a 2-phase scheduling. RPDS differs from existing solutions in several aspects, e.g., the processing elements are heterogeneous, fully-connected, and multi-threading, the application is decomposed into directed acyclic graph tasks with continuous data-packets, and scheduling is conducted at both of initialization and run-time. Experimental results showed our algorithm could increase the largest average throughput by about 30% than those without dynamic phase replication.

Keywords

scheduling network processors task replication partial dynamic scheduling directed acyclic graph 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Malloy, B.A, Lloyd, E.L., Soffa, M.L.: Scheduling DAG’s for Asynchronous Multiprocessor Execution. IEEE Trans. Parallel and Distributed Systems 5(5), 498–508 (1994)CrossRefGoogle Scholar
  2. 2.
    Yan, S., Zhou, X., Wang, L., Wang, H.: GA-Based Automated Task Assignment on Network Processors. In: ICPADS 2005. Proc. of the 11th international Conference on Parallel and Distributed Systems, July 20–22, 2005, pp. 112–118. IEEE Computer Society Press, Los Alamitos (2005)Google Scholar
  3. 3.
    Franklin, M., Datar, S.: Pipeline task scheduling on network processors. In: Proc. of Third Network Processor Workshop in conjunction with Tenth International Symposium on High Performance Computer Architecture (HPCA-10), pp. 103–119 (February 2004)Google Scholar
  4. 4.
    Yang, L., Gohad, T., Ghosh, P., Sinha, D., Sen, A., Richa, A.: Resource mapping and scheduling for heterogeneous network processor systems. In: ANCS 2005. Proc. of the 2005 Symposium on Architecture for Networking and Communications Systems, pp. 19–28 (2005)Google Scholar
  5. 5.
    Ramaswamy, R., Weng, N., Wolf, T.: Application Analysis and Resource Mapping for Heterogeneous Network Processor Architectures. In: Proc. of Network Processor Workshop, Madrid, Spain, pp. 103–119 (2004)Google Scholar
  6. 6.
    Wolf, T., Pappu, P., Franklin, M.A.: Predictive scheduling of network processors. Comput. Networks 41(5), 601–621 (2003)zbMATHCrossRefGoogle Scholar
  7. 7.
    Weng, N., Wolf, T.: Pipelining vs. Multiprocessors-choosing the Right Network Processor System Topology. In: Proc. of ANCHOR 2004, Munich, Germany (2004)Google Scholar
  8. 8.
    Weng, N., Wolf, T.: Profiling and mapping of parallel workloads on network processors. In: Proc. of 20th ACM Symposium on Applied Computing (SAC) (March 2005)Google Scholar
  9. 9.
    Wolf, T., Weng, N., Tai, C.: Design considerations for network processor operating systems. In: ANCS 2005. Proc. of the 2005 Symposium on Architecture for Networking and Communications Systems, October 26–28, 2005, pp. 71–80. ACM Press, New York (2005)CrossRefGoogle Scholar
  10. 10.
    Aggarwal, A., Franklin, M.: Instruction Replication for Reducing Delays Due to Inter-PE Communication Latency. IEEE Trans. Comput. 54(12), 1496–1507 (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Zhiyong Yu
    • 1
  • Zhiyi Yang
    • 1
  • Fan Zhang
    • 1
  • Zhiwen Yu
    • 2
  • Tuanqing Zhang
    • 1
  1. 1.School of Computer Science, Northwestern Polytechnical UniversityP.R. China
  2. 2.Academic Center for Computing and Media Studies, Kyoto UniversityJapan

Personalised recommendations