TY - GEN
T1 - A Worker Implementation on Windows OS for User-PC Computing System
AU - Funabiki, Nobuo
AU - Asai, Yusuke
AU - Aoyagi, Yuuki
AU - Kao, Wen Chung
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/3/2
Y1 - 2016/3/2
N2 - Previously, we have proposed and implemented the User-PC computing (UPC) system as a highly parallel computing platform using idling CPU cycles in members' personal computers (PCs) of the same organization such as a laboratory in a university. This implementation for both the UPC master and the UPC worker covers only Linux OS. However, currently a lot of PCs adopt Windows OS. In this paper, we present an implementation of the UPC worker for Windows OS based on the implementation for Linux OS. During the implementation, we realize the CPU usage limitation and the automatic start-up of the worker program which can reduce loads for PC owner, aside from basic functions of master communications and job executions. Through evaluations, we conclude that the CPU usage rate does not exceed the given limitation except for starting time, and the job execution time on Windows is about twice the use of that on Linux.
AB - Previously, we have proposed and implemented the User-PC computing (UPC) system as a highly parallel computing platform using idling CPU cycles in members' personal computers (PCs) of the same organization such as a laboratory in a university. This implementation for both the UPC master and the UPC worker covers only Linux OS. However, currently a lot of PCs adopt Windows OS. In this paper, we present an implementation of the UPC worker for Windows OS based on the implementation for Linux OS. During the implementation, we realize the CPU usage limitation and the automatic start-up of the worker program which can reduce loads for PC owner, aside from basic functions of master communications and job executions. Through evaluations, we conclude that the CPU usage rate does not exceed the given limitation except for starting time, and the job execution time on Windows is about twice the use of that on Linux.
KW - User-PC computing
KW - Windows
KW - implementation
KW - master
KW - parallel computation
KW - worker
UR - http://www.scopus.com/inward/record.url?scp=84964797852&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84964797852&partnerID=8YFLogxK
U2 - 10.1109/CANDAR.2015.17
DO - 10.1109/CANDAR.2015.17
M3 - Conference contribution
AN - SCOPUS:84964797852
T3 - Proceedings - 2015 3rd International Symposium on Computing and Networking, CANDAR 2015
SP - 279
EP - 282
BT - Proceedings - 2015 3rd International Symposium on Computing and Networking, CANDAR 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 3rd International Symposium on Computing and Networking, CANDAR 2015
Y2 - 8 December 2015 through 11 December 2015
ER -