Skip to main content
Springer Nature Link
Log in

Minimizing Hotspot Delay by Fully Utilizing the Link Bandwidth on 2D Mesh with Virtual Cut-Through Switching

  • Conference paper
Parallel Computing Technologies (PaCT 2005)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 3606))

Included in the following conference series:

Abstract

The hotspot seriously degrades the performance of a parallel algorithm but there have not been many methods proposed for this problem. Without modification of mesh topology a reasonable method is fully utilizing all the links of the hotspot node. A new routing method that incorporates both minimal routes and non-minimal routes was proposed and approved with the hotspot traffic patterns. In particular the routing method decide on misrouting without the congestion detection. The routing method requires only little addition of hardware and it is relatively simple.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Pfister, G.F., Norton, V.A.: Hot Spot Contention and Combining in Multistage Interconnection Networks. IEEE Trans. Computers 34, 943–948 (1985)

    Google Scholar 

  2. Sun, Y., Cheng, P., Lin, X.: Barrier Synchronization on Wormhole-routed Networks. IEEE Trans. Parallel and Distributed Systems 12, 583–597 (2001)

    Article  Google Scholar 

  3. Hyder, R.S., Wood, D.A.: Synchronization Hardware for Networks of Workstations: Performance v.s. cost. In: ACM ICS 1996, pp. 245–252 (1996)

    Google Scholar 

  4. Duato, J., Robles, A., Silla, F.: A Comparison of Router Architectures for Virtual Cut-Through and Wormhole Switching in a NOW Environment. In: 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing, pp. 240–248 (1999)

    Google Scholar 

  5. Lee, H.S., Kim, H.W., Kim, J., Lee, S.: Adaptive Virtual Cut-through as an Alternative to Wormhole Routing. In: Proceedings of the 24th International Conference on Parallel Processing, vol. I, pp. 68–75 (1995)

    Google Scholar 

  6. Ni, L.M., Qiao, W., Yang, M.: Switches and Switch Interconnects. In: 4th International Conference on Massively Parallel Processing Using Optical Interconnections, pp. 122–130 (1997)

    Google Scholar 

  7. Upadhyay, J.H., Varavithya, V., Mohapatra, P.: Efficient and Balanced Adaptive Routing in Two-Dimensional Meshes. In: 1st IEEE Symposium on High-Performance Computer Architecture, pp. 112–121 (1995)

    Google Scholar 

  8. Chuang, P.-J., Chen, J.-T., Jiang, Y.-T.: Balancing Buffer Utilization in Meshes Using a “Restricted Area” Concept. IEEE Trans. Parallel and Distributed Systems 13, 814–827 (2002)

    Article  Google Scholar 

  9. Pifarre, G.D., Gravano, L., Felperin, S.A., Sanz, J.L.C.: Fully Adaptive Minimal Deadlock-Free Packet Routing in Hypercubes, Meshes, and Other Networks: Algorithms and Simulations. IEEE Trans. Parallel and Distributed Systems 5, 247–263 (1994)

    Article  Google Scholar 

  10. Basak, D., Panda, D.K.: Alleviating Consumption Channel Bottleneck in Wormhole-routed k-ary n-cube Systems. IEEE Trans. Parallel and Distributed Systems 9, 481–496 (1998)

    Article  Google Scholar 

  11. Bolding, K., Fulgham, M., Snyder, L.: The Case for Chaotic Adaptive Routing. IEEE Trans. Computers 46, 1281–1997 (1997)

    Article  Google Scholar 

  12. Thottethodi, M.S., Lebeck, A.R., Mukherjee, S.: BLAM: A High-Performance Routing Algorithm for Virtual Cut-Through Networks. In: Proceedings of the International Parallel and Distributed Processing Symposium (2003)

    Google Scholar 

  13. Nguyen, T., Snyder, L.: Performance of Minimal Adaptive Routers. In: Proc. Parallel Computer Routing and Comm. Workshop, pp. 60–71 (1994)

    Google Scholar 

  14. Pifarre, G.D., Gravano, L., Denicolay, G., Sanz, J.L.C.: Adaptive Deadlock- and Livelock-Free Routing in the Hypercube Network. IEEE Trans. Parallel and Distributed Systems 5, 1121–1139 (1994)

    Article  Google Scholar 

  15. Dally, W.J.: Performance Analysis of k-ary n-cube Interconnection Networks. IEEE Trans. Computers 39(6), 775–785 (1990)

    Article  Google Scholar 

  16. Greenberg, A.G., Hajek, B.: Deflection Routing in Hypercube Networks. IEEE Trans. Communications COM-40, 1070–1081 (1992)

    Article  Google Scholar 

  17. Konstantinidou, S., Snyder, L.: The Chaos Router. IEEE Trans. Computers 43, 1386–1397 (1994)

    Article  Google Scholar 

  18. Ok, M.-H., Park, M.-S.: A Novel Collective Communication Scheme on Packet-switched 2-D Mesh Interconnection. In: Proc. Int’l Conf. VECPAR 2000, pp. 147–158 (2000)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul, 136-701, South Korea

    MinHwan Ok & Myong-soon Park

Authors
  1. MinHwan Ok
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Myong-soon Park
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Editors and Affiliations

  1. Novosibirsk State Technical University,  

    Victor Malyshkin

Rights and permissions

Reprints and permissions

Copyright information

© 2005 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Ok, M., Park, Ms. (2005). Minimizing Hotspot Delay by Fully Utilizing the Link Bandwidth on 2D Mesh with Virtual Cut-Through Switching. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2005. Lecture Notes in Computer Science, vol 3606. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11535294_22

Download citation

  • DOI: https://doi.org/10.1007/11535294_22

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-28126-9

  • Online ISBN: 978-3-540-31826-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics