Abstract
Wormhole routing techniques are widely used as a low-latency message passing mechanism in parallel computers and System Area Networks (SAN). SAN is the new high-performance computing trend, providing a low cost-performance ratio. Server Net (Compaq) and Myrinet are providing COTS (routers, PCI interface cards and software drivers) for building high-speed scalable SAN, based on wormhole routing techniques. Different regular topologies can be built with COTS. One of them is the 2D-torus network topology. In this paper a fault-tolerant adaptive wormhole routing algorithm for the 2D-torus is proposed. The algorithm uses two virtual networks North Last and Torus North (NL/TN) and is deadlock-and livelock-free. In a fault-free torus the method provides minimum path routing for 97% of all source-destination pairs. The algorithm can tolerate up to N-1 node and link faults in a N × N 2D torus. In the connected 2D torus all fault-free nodes are reachable. The algorithm is decentralized and does not require global knowledge about the status of the networks. The simulation results present the effect of the network size, the request generation rate, the number of faults, and the fault pattern on the performance characteristics — throughput, latency and saturation point.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Acosta, J. R. and Avresky, D. R. (1999). Fault-Tolerant Routing in 2D Torus with Wormhole Routing. In Proceedings of PDPTA ′99 International Conference, volume V, pages 2483–2487.
Avresky, D. R., Cunningham, C. M., and Ravichandran, H. (1999). Fault-Tolerant Routing for Wormhole-Routed Two-Dimensional Meshes. Int. Journal of Computer Systems Science&Engineering, 14(6).
Avresky, D. R., Shurbanov, V., and Horst, R. (1998). The Effect of the Router Arbitration Policy on the Scalability of ServerNet™ Topologies. J. of Microprocessors and Microsystems, 21(9):546–561. Elsevier Science, The Netherlands.
Baker, W. E., Horst, R. W., Sonnier, D. P., and Watson, W. J. (1995). A Flexible ServerNet-based Fault-Tolerant Architecture. In Proc. of the 25th Int. Symp. Fault-Tolerant Computing, pages 2–11, Pasadena, CA, U.S.A.
Boppana, R. V. and Chalasani, S. (1995). Fault-Tolerant Wormhole Routing Algorithms for Mesh Networks. IEEE Transactions on Computers, 44(7):848–863.
Boura, Y. M. and Das, C. R. (1995). Fault-tolerant Routing in Mesh Networks. Urbana-Champaign, Illinois, U.S.A.
Cunningham, C. and Avresky, D. R. (1995). Fault-Tolerant Routing for Worm-hole Routed Two-Dimensional Meshes. In IEEE First International Symposium on High Performance Computer Architecture, pages 122–131.
Dally, W. J. and Seitz, C. L. (1987). Deadlock-Free Message Routing in Multiprocessor. IEEE Trans. on Computers, pages 547–553.
Daniel H. Linder, J. C. H. (1991). An Adaptive and Fault Tolerant Worm-hole Routing Strategy for k-ary n-cubes. IEEE Transactions on Computers, 40(1):2–12.
Glass, C. J. and Ni, L. M. (1992). The Turn Model for Adaptive Routing. Proceedings of the 19th Annual International Symposium on Computer Architecture, pages 278–286.
Glass, C. J. and Ni, L. M. (1993). Fault-Tolerant Wormhole Routing in Meshes. In Proceedings of the 23rd International Symposium on Fault-Tolerant Computing, pages 240–249, Toulouse, France.
Heirich, A., Garcia, D., Knowles, M., and Horst, R. (1998). ServerNet-II: A reliable interconnet for scalable high-performance cluster computing. Compaq Computer Corp., Tandem Division.
Hilland, J. (1998). Virtual Interface Architecture Specification. Tandem Computers Inc.
Horst, R. W. (1995). TNet: A Reliable System Area Network. IEEE Micro, pages 37–45.
Kim, J. and Shin, K. G. (1993). Deadlock-Free Fault-Tolerant Routing in Injured Hypercubes. IEEE Trans. on Computers, 42(9):1078–1088.
Lee, T. C. and Hayes, J. R (1992). A Fault-Tolerant Communication Scheme for Hypercube Computers. IEEE Trans. on Computers, 41(10):1242–1255.
Luis Gravano, Gustavo D. Pifarre, P. E. B. and Sanz, J. L. (1994). Adaptive Deadlock-and Livelock-Free Routing With all Minimal Paths in Torus Networks. IEEE Transactions on Parallel and Distributed Systems, 5(12):1233–1251.
McCaffrey, Z. and Avresky, D. R. (1998). Deadlock and Livelock-Free Worm-hole Routing for 2-D Toruses. In Proceedings of the PDPTA′98, Las Vegas, NV, USA.
Ni, L. M. and McKinley, P. K. (1993). A Survey of Wormhole Routing Techniques in Direct Networks. Computer, 26(2):62–76.
Suh, Y., Dao, B. V., Duato, J., and Yalamanchili, S. (1995). Software based fault-tolerant oblivios routing in pipelined networks. Urbana-Champaign, Illinois, U.S.A.
VI Software Development Team (Tandem, A. C. C. (1998). Server Net I Virtual Interface Software External Specification. Tandem Computers Inc.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer Science+Business Media New York
About this chapter
Cite this chapter
Avresky, D., Acosta, J., Shurbanov, V., McAffrey, Z. (2000). Adaptive Minimal-Path Routing in 2-Dimensional Torus Servernet San. In: Avresky, D.R. (eds) Dependable Network Computing. The Springer International Series in Engineering and Computer Science, vol 538. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-4549-1_11
Download citation
DOI: https://doi.org/10.1007/978-1-4615-4549-1_11
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-7053-6
Online ISBN: 978-1-4615-4549-1
eBook Packages: Springer Book Archive