Faculty of Information Technology, BUT

Publication Details

An Evolutionary Approach to Collective Communication Scheduling

JAROŠ Jiří, OHLÍDAL Miloš and DVOŘÁK Václav. An Evolutionary Approach to Collective Communication Scheduling. In: 2007 Genetic and Evolutionary Computation Conference. Volume II. New York: Association for Computing Machinery, 2007, pp. 2037-2044. ISBN 978-1-59593-697-4.
Czech title
Evoluční přístup k plánování kolektivních komunikací
Type
conference paper
Language
english
Authors
URL
Keywords
Collective communications, communication scheduling, evolutionary optimization.
Abstract
In this paper, we describe two evolutionary algorithms aimed at scheduling collective communications on interconnection networks of parallel computers. To avoid contention for links and associated delays, collective communications proceed in synchronized steps. Minimum number of steps is sought for the given network topology, wormhole (pipelined) switching, minimum routing and given sets of sender and/or receiver nodes. Used algorithms are able not only re-invent optimum schedules for known symmetric topologies like hyper-cubes, but they can find schedules even for any asymmetric or irregular topologies in case of general many-to-many collective communications. In most cases does the number of steps reach the theoretical lower bound for the given type of collective communication; if it does not, non-minimum routing can provide further improvement. Optimum schedules may serve for writing high-performance communication routines for application-specific networks on chip or for development of communication libraries in case of general-purpose interconnection networks.
Published
2007
Pages
2037-2044
Proceedings
2007 Genetic and Evolutionary Computation Conference
Series
Volume II
Conference
Genetic and Evolutionary Computation Conference , London, GB
ISBN
978-1-59593-697-4
Publisher
Association for Computing Machinery
Place
New York, US
BibTeX
@INPROCEEDINGS{FITPUB8369,
   author = "Ji\v{r}\'{i} Jaro\v{s} and Milo\v{s} Ohl\'{i}dal and V\'{a}clav Dvo\v{r}\'{a}k",
   title = "An Evolutionary Approach to Collective Communication Scheduling",
   pages = "2037--2044",
   booktitle = "2007 Genetic and Evolutionary Computation Conference",
   series = "Volume II",
   year = 2007,
   location = "New York, US",
   publisher = "Association for Computing Machinery",
   ISBN = "978-1-59593-697-4",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/8369"
}
Back to top