DocumentCode :
258292
Title :
Modeling, optimization and performance prediction of parallel algorithms
Author :
Hudik, Martin ; Hodon, Michal
Author_Institution :
Dept. of Tech. Cybern., Univ. of Zilina, Zilina, Slovakia
fYear :
2014
fDate :
23-26 June 2014
Firstpage :
1
Lastpage :
7
Abstract :
The high intensity of research and modeling in fields of mathematics, physics, biology and chemistry requires new computing resources. For the big computational complexity of such tasks computing time is large and costly. The most efficient way to increase efficiency is to adopt parallel principles. Purpose of this paper is to present the issue of parallel computing with emphasis on the analysis of parallel systems, the impact of communication delays on their efficiency and on overall execution time. Paper focuses is on finite algorithms for solving systems of linear equations, namely the matrix manipulation (Gauss elimination method GEM). Algorithms are designed for architectures with shared memory (openMP), distributed-memory (MPI) and for their combination (MPI+openMP). The properties of the algorithms were analytically determined and they were experimentally verified. The conclusions are drawn for theory and practice.
Keywords :
Gaussian processes; application program interfaces; computational complexity; mathematics computing; matrix algebra; message passing; parallel algorithms; shared memory systems; GEM; Gauss elimination method; MPI+openMP; biology; chemistry; communication delays; computational complexity; distributed-memory; finite algorithms; linear equations; mathematics; matrix manipulation; parallel algorithms; performance prediction; physics; shared memory; tasks computing time; Algorithm design and analysis; Computational modeling; Computer architecture; Computers; Equations; Mathematical model; Parallel algorithms; Gauss elimination method; collective communication operations; efficiency; modeling; parallel algorithms; parallel architecture; parallel computation; performance prediction; pipelined broadcast; system of linear equations;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computers and Communication (ISCC), 2014 IEEE Symposium on
Conference_Location :
Funchal
Type :
conf
DOI :
10.1109/ISCC.2014.6912632
Filename :
6912632
Link To Document :
بازگشت