Avoiding communication hot-spots in interconnection networks

In creating interconnection networks for parallel computers, an efficient design is crucial because of its impact on performance. A high-speed routing scheme that minimises contention and avoids the formation of hot-spots should be included. We have developed a new method to uniformly balance communication traffic over the interconnection network, called distributed routing balancing (DRB), based on limited and load-controlled path expansion in order to maintain low message latency. DRB defines how to create alternative paths to expand single paths (expanded path definition) and when to use them depending on traffic load (expanded path selection). An evaluation in terms of latency and bandwidth is presented. Some conclusions and comparisons with existing methods are given. It is demonstrated that DRB is a method to effectively balance network traffic.

[1]  J DallyWilliam,et al.  Performance Analysis of k-ary n-cube Interconnection Networks , 1990 .

[2]  Leslie G. Valiant,et al.  Universal schemes for parallel communication , 1981, STOC '81.

[3]  Lionel M. Ni,et al.  The Turn Model for Adaptive Routing , 1992, [1992] Proceedings the 19th Annual International Symposium on Computer Architecture.

[4]  Shahid H. Bokhari,et al.  Multiphase complete exchange on Paragon, SP2, and CS-2 , 1996, IEEE Parallel Distributed Technol. Syst. Appl..

[5]  Gregory F. Pfister,et al.  “Hot spot” contention and combining in multistage interconnection networks , 1985, IEEE Transactions on Computers.

[6]  Agustin Arruabarrena,et al.  Optimal Distance Networks of Low Degree for Parallel Computers , 1991, IEEE Trans. Computers.

[7]  Lionel M. Ni,et al.  The turn model for adaptive routing , 1992, ISCA '92.

[8]  Mike Hillyard Asynchronous Transfer Mode , 1993 .

[9]  Emilio Luque,et al.  Improving parallel computer communication: dynamic routing balancing , 1998, Proceedings of the Sixth Euromicro Workshop on Parallel and Distributed Processing - PDP '98 -.

[10]  José Duato,et al.  A New Theory of Deadlock-Free Adaptive Routing in Wormhole Networks , 1993, IEEE Trans. Parallel Distributed Syst..

[11]  S. Konstantinidou,et al.  Chaos router: architecture and performance , 1991, [1991] Proceedings. The 18th Annual International Symposium on Computer Architecture.

[12]  William J. Dally Virtual-channel flow control , 1990, ISCA '90.

[13]  Anant Agarwal,et al.  Limits on Interconnection Network Performance , 1991, IEEE Trans. Parallel Distributed Syst..

[14]  Martin de Prycker,et al.  Asynchronous Transfer Mode, Solution for Broadband Isdn , 1991 .

[15]  Andrew A. Chien,et al.  Planar-adaptive routing: low-cost adaptive networks for multiprocessors , 1992, ISCA '92.

[16]  William J. Dally,et al.  Deadlock-Free Message Routing in Multiprocessor Interconnection Networks , 1987, IEEE Transactions on Computers.

[17]  William J. Dally,et al.  Performance Analysis of k-Ary n-Cube Interconnection Networks , 1987, IEEE Trans. Computers.

[18]  Marc Snir,et al.  The Communication Software and Parallel Environment of the IBM SP2 , 1995, IBM Syst. J..

[19]  Andrew A. Chien,et al.  Compressionless Routing: A Framework for Adaptive and Fault-Tolerant Routing , 1997, IEEE Trans. Parallel Distributed Syst..

[20]  Sudhakar Yalamanchili,et al.  Interconnection Networks: An Engineering Approach , 2002 .

[21]  P. H. Welch,et al.  Networks, Routers and Transputers: Function, Performance and Applications , 1993 .

[22]  J. Nunamaker,et al.  Proceedings of the 32nd Hawaii International Conference on System Sciences , 1999 .