Advertisement

FROOTS – Fault Handling in Up*/Down* Routed Networks with Multiple Roots

  • Ingebjørg Theiss
  • Olav Lysne
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2913)

Abstract

Faults in irregular networks have previously been handled through a global or local network reconfiguration, which leaves the network unavailable or disconnected until the reconfiguration is finished. We present FRoots, which leaves the network available and connected by using redundant paths. Only packets residing in the network at the fault time might become unroutable, and FRoots can be designed to offer an alternative path for such packets. To accommodate sequential, multiple faults, reconfiguration of the obsolete paths is done without a severe downgrade of network throughput. FRoots is based on a load balancing version of Up*/Down* routing, and uses a modest number of virtual channels to achieve redundancy properties. However, FRoots utilizes all channels in the fault-free case, thus experiencing no performance loss compared to its predecessor routing algorithm.

Keywords

Span Tree Leaf Node Fault Tolerance Virtual Channel Articulation Point 
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.
    InfiniBand Trade Association, InfiniBand Architecture SpecificationGoogle Scholar
  2. 2.
    Duato, J., Yalamanchili, S., Ni, L.: Interconnection Networks: An Engineering Approach. IEEE, Los Alamitos (1997)Google Scholar
  3. 3.
    Schröder, M.D., et al.: Autonet: a High-Speed, Self-Configuring Local Area Network Using Point-to-point Links. SRC Research Report 59 (December 1990)Google Scholar
  4. 4.
    Boden, N.J., et al.: Myrinet–A Gigabit-per-Second Local-Area Network. In: IEEE MICRO (1995)Google Scholar
  5. 5.
    Lysne, O., Skeie, T.: Load Balancing of Irregular System Area Networks through Multiple Roots. In: Int. Conf. on Communications in Computing (2001)Google Scholar
  6. 6.
    Sancho, J.C., Robles, A.: Improving the Up*/Down* Routing Scheme for Networks of Workstations. In: Bode, A., Ludwig, T., Karl, W.C., Wismüller, R. (eds.) Euro-Par 2000. LNCS, vol. 1900, pp. 882–889. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  7. 7.
    Skeie, T., Lysne, O., Theiss, I.: Layered Shortest Path (LASH) Routing in Irregular System Area Networks. In: Proc. of IPDPS 2002, IEEE, Los Alamitos (2002)Google Scholar
  8. 8.
    Sancho, J.C., et al.: Effective Methodology for Deadlock-Free Minimal Routing in InfiniBand Networks. In: Proc. of ICPP, pp. 409–418. IEEE, Los Alamitos (2002)CrossRefGoogle Scholar
  9. 9.
    Chien, A.A., Kim, J.H.: Planar-adaptive routing: Low-cost adaptive networks for multiprocessors. Journal of the ACM 42(1), 91–123 (1995)zbMATHCrossRefGoogle Scholar
  10. 10.
    Theiss, I., Lysne, O.: LORE - Local Reconfiguration for Fault Managment in Arbitrary Networks. Report 2003-02, Simula Research Laboratory, Norway (2003)Google Scholar
  11. 11.
    Casado, R., et al.: Performance evaluation of dynamic reconfiguration in highspeed local area networks. In: Proc. of the 6th Int. Symp. on High-Performance Computer Architecture (2000)Google Scholar
  12. 12.
    Pang, R., Pinkston, T., Duato, J.: The Double Scheme: Deadlock-Free Dynamic Reconfiguration of Cut-Through Networks. In: Proc. of ICPP 2000, IEEE, Los Alamitos (2000)Google Scholar
  13. 13.
    Lysne, O., Duato, J.: Fast Dynamic Reconfiguration in Irregular Networks. In: Proc. of ICPP 2000, pp. 449–458. IEEE, Los Alamitos (2000)CrossRefGoogle Scholar
  14. 14.
    Weiss, M.A.: Data Structures and Algorithm Analysis. Benjamin/Cummings (1992)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Ingebjørg Theiss
    • 1
  • Olav Lysne
    • 1
  1. 1.Simula Research LaboratoryNorway

Personalised recommendations