Advertisement

Entropy Throttling: A Physical Approach for Maximizing Packet Mobility in Interconnection Networks

  • Takashi Yokota
  • Kanemitsu Ootsu
  • Fumihito Furukawa
  • Takanobu Baba
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4186)

Abstract

A large-scale direct interconnection network usually consists of enormous number of simple routers. However, its behavior is sometimes very complicated. Such a complicated behavior prevents us from accurate understanding and efficient control of the network. Among serious problems in interconnection networks, congestion control is of extreme importance since network performance is drastically degraded by a congested situation. We focus our discussion on throttling, injection limitation in other words, as one of the most hopeful solutions to the congestion problem. Our approach is inspired from physics. We define entropy as a desirable metric for representing the network’s congestion level. We also define packet mobility ratio as a proper approximation of entropy. Thus we reach a new throttling method called ‘Entropy Throttling’ that is based on theoretical discussion on congestion. Evaluation results by our simulator reveal effectiveness of the proposed method.

Keywords

Congestion Control Interconnection Network Average Latency Entropy Measure Congestion Level 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Duato, J., Yalamanchili, S., Ni, L.: Interconnection Networks: An Engineering Approach. Morgan Kaufmann Pub., San Francisco (2003)Google Scholar
  2. 2.
    Dally, W.J., Towles, B.: Principles and Practices of Interconnection Networks. Morgan Kaufmann Pub., San Francisco (2004)Google Scholar
  3. 3.
    Yokota, T., Ootsu, K., Furukawa, F., Baba, T.: A Cellular Automata Approach for Large-Scale Interconnection Network Simulation. IEICE Technical Report, CPSY2005–32 (December 2005)Google Scholar
  4. 4.
    Yokota, T., Ootsu, K., Furukawa, F., Baba, T.: A Cellular Automata Approach for Understanding Congestion in Interconnection Networks. IPSJ Trans. Advanced Computing Systems (to appear, 2006)Google Scholar
  5. 5.
    Yokota, T., Ootsu, K., Furukawa, F., Baba, T.: Phase Transition Phenomena in Interconnection Networks of Massively Parallel Computers. Journal of Physical Society of Japan 75(7) (to appear, 2006)Google Scholar
  6. 6.
    Yokota, T., Matsuoka, H., Okamoto, K., Hirono, H., Sakai, S.: Virtual Control Channel and its Application to the Massively Parallel Computer RWC-1. In: Proc. Int. Conf. on High Performance Computing (HiPC 1997), pp. 443–448 (December 1997)Google Scholar
  7. 7.
    Nishitani, M., Ezura, S., Yokota, T., Ootsu, K., Baba, T.: Preliminary Research of a Novel Routing Algorithm Cross-Line Using Dynaic Information. In: Proc. IASTED PDCS 2004, pp. 107–112 (November 2004)Google Scholar
  8. 8.
    Yokota, T., Nishitani, M., Ootsu, K., Furukawa, F., Baba, T.: Cross-Line — A Globally Adaptive Control Method of Interconnection Network. In: Proc. ISHPC–VI (September 2005)Google Scholar
  9. 9.
    Yokota, T., Ootsu, K., Furukawa, F., Baba, T.: Cross-Line: A Novel Routing Algorithm that Uses Global Information. IPSJ Trans. Advanced Computing Systems 46 SIG 16 (ACS 12), 28–42 (2005)Google Scholar
  10. 10.
    Bernstein, H.J.: Some Comments on Highly Dynamic Network Routing. Tech. Rep. 371, Computer Science Dept., New York Univ. (May 1988)Google Scholar
  11. 11.
    Baydal, E., López, P., Duato, J.: A Family of Mechanisms for Congestion Control in Wormhole Networks. IEEE Trans. Parallel and Distributed Systems 16(9), 772–784 (2005)CrossRefGoogle Scholar
  12. 12.
    López, P., Martínez, J.M., Duato, J.: DRIL: Dynamically Reduced Message Injection Limitation Mechanism for Wormhole Networks. In: Proc. ICPP 1998, pp. 535–562 (1998)Google Scholar
  13. 13.
    Obaidat, M.S., Al-Awwami, Z.H., Al-Mulhem, M.: A new injection limitation mechanism for wormhole networks. Computer Communications 25, 997–1008 (2002)CrossRefGoogle Scholar
  14. 14.
    Thottethodi, M., Lebeck, A.R., Mukherjee, S.S.: Self-Tuned Congestion Control for Multiprocessor Networks. In: Proc. HPCA–7, pp. 107–118 (2001)Google Scholar
  15. 15.
    Thottethodi, M., Lebeck, A.R., Mukherjee, S.S.: Exploiting Global Knowledge to Achieve Self-Tuned Congestion Sontrol for k-Ary n-Cube Networks. IEEE Trans. Parallel and Distributed Systems 15(3), 257–272 (2004)CrossRefGoogle Scholar
  16. 16.
    Baydal, E., López, P., Duato, J.: A Congestion Control Mechanism for Wormhole Networks. In: Proc. 9th Euromicro Workshop on Parallel and Distributed Processing (EUROPDP 2001), pp. 19–26 (February 2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Takashi Yokota
    • 1
  • Kanemitsu Ootsu
    • 1
  • Fumihito Furukawa
    • 2
  • Takanobu Baba
    • 1
  1. 1.Department of Information ScienceUtsunomiya UniversityTochigiJapan
  2. 2.Learning Technology LaboratoryTeikyo UniversityTochigiJapan

Personalised recommendations