Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/52879
Title: Anticipating performance information of newly portable computers on the WLAN for load balancing
Authors: Sanchez, D 
Macias, EM 
Suarez, A 
Issue Date: 2004
Publisher: 0302-9743
Journal: Lecture Notes in Computer Science 
Conference: 5th International Conference on Parallel Processing and Applied Mathematics 
5th International Conference on Parallel Processing and Applied Mathematics, PPAM 2003 
Abstract: Parallel and distributed computing over heterogeneous networks is a hot topic nowadays. We have used a combination of a Local Area Network (LAN) and a Wireless LAN (WLAN) to implement Master-Slave parallel distributed programs which exhibit dependences among iterations and parallel calculation inside an iteration. In this environment it is necessary to consider the unbalance communication bandwidths and computing power of fixed and portable computers to avoid long idle times. To solve this problem, in our previous work, we developed an efficient load balancing strategy considering that no new processes are spawned in new portable computers at run time. Due to it is not realistic, in this paper we present a new method for balancing the parallel calculations when parallel processes are spawned on portable computers which uses performance information supplied by a framework based on the standard Simple Network Management Protocol (SNMP). In this way, idle times are minimized and better execution times are obtained.
URI: http://hdl.handle.net/10553/52879
ISBN: 3-540-21946-3
9783540219460
ISSN: 0302-9743
Source: Parallel Processing And Applied Mathematics[ISSN 0302-9743],v. 3019, p. 946-953
Appears in Collections:Actas de congresos
Show full item record

SCOPUSTM   
Citations

3
checked on Sep 15, 2024

WEB OF SCIENCETM
Citations

2
checked on Feb 25, 2024

Page view(s)

39
checked on Feb 17, 2024

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.