Detail publikace

Design of Low-Cost Communication Algorithms for Irregular Networks

Originální název

Design of Low-Cost Communication Algorithms for Irregular Networks

Anglický název

Design of Low-Cost Communication Algorithms for Irregular Networks

Jazyk

en

Originální abstrakt

Communication between non-adjacent nodes in irregular interconnection networks mostly relies on routing tables. Efficient implementation of group communication algorithms such as total exchange also needs similar tables. Since such tables cannot be derived mathematically, we have used a parallel genetic algorithm GAroute to design a sub-optimal routing schedule for a sequence of communication steps (neighbor-to-neighbor transfers) during group communications. Genetic evolution does not necessarily lead to the minimum number of steps, but provides satisfactory low-cost communication algorithms. Accuracy of obtained results has been tested with a regular hypercube topology with known complexity of communication.

Anglický abstrakt

Communication between non-adjacent nodes in irregular interconnection networks mostly relies on routing tables. Efficient implementation of group communication algorithms such as total exchange also needs similar tables. Since such tables cannot be derived mathematically, we have used a parallel genetic algorithm GAroute to design a sub-optimal routing schedule for a sequence of communication steps (neighbor-to-neighbor transfers) during group communications. Genetic evolution does not necessarily lead to the minimum number of steps, but provides satisfactory low-cost communication algorithms. Accuracy of obtained results has been tested with a regular hypercube topology with known complexity of communication.

BibTex


@inproceedings{BUT16907,
  author="Jiří {Staroba} and Václav {Dvořák}",
  title="Design of Low-Cost Communication Algorithms for Irregular Networks",
  annote="Communication between non-adjacent nodes in irregular interconnection networks mostly relies on routing tables. Efficient implementation of group communication algorithms such as total exchange also needs similar tables. Since such tables cannot be derived mathematically, we have used a parallel genetic algorithm GAroute to design a sub-optimal routing schedule for a sequence of communication steps (neighbor-to-neighbor transfers) during group communications. Genetic evolution does not necessarily lead to the minimum number of steps, but provides satisfactory low-cost communication algorithms. Accuracy of obtained results has been tested with a regular hypercube topology with known complexity of communication.",
  address="University of Haute Alsace",
  booktitle="Proceedings of the 3rd International Conference on Networking ICN '04",
  chapter="16907",
  institution="University of Haute Alsace",
  year="2004",
  month="march",
  pages="980--985",
  publisher="University of Haute Alsace",
  type="conference paper"
}