Detail publikace

An Evolutionary Design Technique for Collective Communications on Optimal Diameter-Degree Networks

Originální název

An Evolutionary Design Technique for Collective Communications on Optimal Diameter-Degree Networks

Anglický název

An Evolutionary Design Technique for Collective Communications on Optimal Diameter-Degree Networks

Jazyk

en

Originální abstrakt

Scheduling collective communications (CC) in networks based on optimal graphs and digraphs has been done with the use of the evolutionary techniques. Inter-node communication patterns scheduled in the minimum number of time slots have been obtained. Numerical values of communication times derived for illustration can be used to estimate speedup of typical applications that use CC frequently. The results show that evolutionary techniques often lead to ultimate scheduling of CC that reaches theoretical bounds on the number of steps. Analysis of fault tolerance by the same techniques revealed graceful CC performance degradation for a single link fault. Once the faulty link is located, CC can be re-scheduled during a recovery period.

Anglický abstrakt

Scheduling collective communications (CC) in networks based on optimal graphs and digraphs has been done with the use of the evolutionary techniques. Inter-node communication patterns scheduled in the minimum number of time slots have been obtained. Numerical values of communication times derived for illustration can be used to estimate speedup of typical applications that use CC frequently. The results show that evolutionary techniques often lead to ultimate scheduling of CC that reaches theoretical bounds on the number of steps. Analysis of fault tolerance by the same techniques revealed graceful CC performance degradation for a single link fault. Once the faulty link is located, CC can be re-scheduled during a recovery period.

BibTex


@inproceedings{BUT30492,
  author="Jiří {Jaroš} and Václav {Dvořák}",
  title="An Evolutionary Design Technique for Collective Communications on Optimal Diameter-Degree Networks",
  annote="Scheduling collective communications (CC) in networks based on optimal graphs and
digraphs has been done with the use of the evolutionary techniques. Inter-node
communication patterns scheduled in the minimum number of time slots have been
obtained. Numerical values of communication times derived for illustration can be
used to estimate speedup of typical applications that use CC frequently. The
results show that evolutionary techniques often lead to ultimate scheduling of CC
that reaches theoretical bounds on the number of steps. Analysis of fault
tolerance by the same techniques revealed graceful CC performance degradation for
a single link fault. Once the faulty link is located, CC can be re-scheduled
during a recovery period.",
  address="Association for Computing Machinery",
  booktitle="2008 Genetic and Evolutionary Computational Conference GECCO",
  chapter="30492",
  edition="NEUVEDEN",
  howpublished="print",
  institution="Association for Computing Machinery",
  year="2008",
  month="july",
  pages="1539--1546",
  publisher="Association for Computing Machinery",
  type="conference paper"
}