Cluster Computing

, Volume 10, Issue 2, pp 229–239 | Cite as

A queueing model for predicting message latency in uni-directional k-ary n-cubes with deterministic routing and non-uniform traffic

Article

Abstract

The interconnection network is one of the key architectural components in any parallel computer. The distribution of the traffic injected into the network is among the factors that greatly influences network performance. The uniform traffic pattern has been adopted in many existing network performance evaluation studies due to the tractability of the resulting analytical modelling approach. However, many real applications exhibit non-uniform traffic patterns such as hot-spot traffic. K-ary n-cubes have been the mostly widely used in the implementation of practical parallel systems. Extensive research studies have been conducted on the performance modelling and evaluation of these networks. Nonetheless, most of these studies have been confined to uniform traffic distributions and have been based on software simulation. The present paper proposes a new stochastic model to predict message latency in k-aryn-cubes with deterministic routing in the presence of hot-spot traffic. The model has been validated through simulation experiments and has shown a close agreement with simulation results.

Keywords

Interconnection networks k-ary n-cubes Deterministic routing Hot spot Performance modelling M/G/1 queueing systems 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Agrawal, A.: Limits on interconnection network performance. IEEE Trans. Parallel Distributed Syst. 2(4), 398–412 (1991) CrossRefGoogle Scholar
  2. 2.
    Chien, A.A.: A cost and speed model for k-ary n-cube wormhole routers. IEEE Trans. Parallel Distributed Syst. 9(2), 150–162 (1998) CrossRefMathSciNetGoogle Scholar
  3. 3.
    Cray Research Inc.: The Cray T3E scalable parallel processing system, On Cray’s Web http://www.cray.com/PUBLIC/productinfo/T3E/CRAY_T3E.html
  4. 4.
    Dally, W.J.: Virtual channel flow control. IEEE Trans. Parallel Distributed Syst. 3(2), 194–205 (1992) CrossRefGoogle Scholar
  5. 5.
    Dally, W.: Performance analysis of k-ary n-cubes interconnection network. IEEE Trans. Comput. 39(6), 775–785 (1990) CrossRefGoogle Scholar
  6. 6.
    Dally, W.J., Seitz, C.: Deadlock-free message routing in multiprocessor interconnection networks. IEEE Trans. Comput. 36(5), 547–553 (1987) MATHGoogle Scholar
  7. 7.
    Draper, J.T., Ghosh, J.: A comprehensive analytical model for wormhole routing in multicomputer systems. J. Parallel Distributed Comput. 32(2), 202–214 (1994) CrossRefGoogle Scholar
  8. 8.
    Duato, J.: A new theory of deadlock-free adaptive routing in wormhole-routed networks. IEEE Trans. Parallel Distributed Syst. 4(12), 1320–1331 (1993) CrossRefGoogle Scholar
  9. 9.
    Fillo, M., Keckler, S.W., Dally, W.J., Carter, N.P., Chang, A., Gurevich, Y., Lee, W.S.: The M-Machine multicomputer. J. Parallel Program. 25(3), 183–212 (1997) Google Scholar
  10. 10.
    Greenberg, R., Guan, L.: Modelling and comparison of wormhole-routed mesh and torus networks. In: Proceedings of the 9th Int. Conf. Parallel and Distributed Computing and Systems, 1997 Google Scholar
  11. 11.
    Kessler, R.E., Swarszmeier, J.L.: Cray T3D: A new dimension for Cray Research. In: Proceedings Compcon, 1993, pp. 176–182 Google Scholar
  12. 12.
    Kleinrock L.: Queueing Syst., vol. 1. Wiley, New York (1975) Google Scholar
  13. 13.
    Laudon, J., Lenoski, D.: The SGI Origin a ccNUMA highly scalable server. In: Proceedings of the ACM/IEEE 24th Int. Symp. Computer Architecture, 1997, pp. 241–251 Google Scholar
  14. 14.
    Loucif, S., Ould-Khaoua, M.: Modelling latency in deterministic wormhole-routed hypercubes under hot-spot traffic. J. Supercomput. 27(3), 265–278 (2004) MATHCrossRefGoogle Scholar
  15. 15.
    Loucif, S., Ould-Khaoua, M.: Analysis of fully adaptive routing in wormhole-routed tori. Parallel Comput. 27(1), 1477–1487 (1999) CrossRefMathSciNetGoogle Scholar
  16. 16.
    Min, G., Ould-Khaoua, M.: A new performance model for wormhole-switched k-ary n-cubes. IEEE Trans. Comput. 53(5), 601–613 (2004) CrossRefGoogle Scholar
  17. 17.
    Ni, L.M., McKinley, K.: A Survey of wormhole routing techniques in direct networks. IEEE Comput. 26, 62–76 (1993) Google Scholar
  18. 18.
    Noakes, M., et al.: The J-Machine multicomputer: An architectural evaluation. In: Proceedings of the 20th Int. Symp. Computer Architecture, 1993 Google Scholar
  19. 19.
    N Cube-2, N CUBE Company, N CUBE 6400 Processor Manual (1990) Google Scholar
  20. 20.
    Ould-Khaoua, M., Sarbazi-Azad, H.: An analytical model of adaptive wormhole routing in hypercubes in the presence of hot-spot traffic. IEEE Trans. Parallel Distributed Syst. 12(3), 283–288 (2001) CrossRefMathSciNetGoogle Scholar
  21. 21.
    Ould-Khaoua, M.: A performance model for Duato’s adaptive routing algorithm in k-ary n-cubes. IEEE Trans. Comput. 48(12), 1–8 (1999) CrossRefMathSciNetGoogle Scholar
  22. 22.
    Peterson, C.J., Sutton, J., Wiley, P.: iWARP: A 100-MPOS LIW microprocessor for multicomputers. IEEE Micro 11(13), 26–87 (1991) CrossRefGoogle Scholar
  23. 23.
    Pfister, G.J., Norton, V.A.: Hot-spot contention and combining in multistage interconnection networks. IEEE Trans. Comput. 34(10), 943–948 (1985) Google Scholar
  24. 24.
    Sarbazi-Azad, H., Mackenzie, L., Ould-Khaoua, M.: Analytical modelling of wormhole-routed k-ary n-cubes in the presence of hot-spot traffic. IEEE Trans. Comput. 50(7), 623–634 (2001) CrossRefMathSciNetGoogle Scholar
  25. 25.
    Vaidya, A.S., et al.: Impact of virtual channels and adaptive routing on application performance. IEEE Trans. Parallel Distributed Syst. 12(2), 223–237 (2001) CrossRefGoogle Scholar
  26. 26.
    Xu, H., et al.: Efficient implementation of barrier synchronisation in wormhole-routed hypercube multicomputers. J. Parallel Distributed Comput. 16, 172–184 (1992) CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  1. 1.Department of Computing ScienceUniversity of GlasgowGlasgowUK
  2. 2.Department of Electrical & Computer EngineeringSultan Qaboos UniversityAl-KhodOman
  3. 3.Department of ComputingUniversity of BradfordBradfordUK

Personalised recommendations