DMCA
Experiments with mpC: Efficient Solving Regular Problems on Heterogeneous Networks of Computers via Irregularization
BibTeX
@MISC{Irregularization_experimentswith,
author = {Via Irregularization and Dmitry Arapov and Alexey Kalinov and Alexey Lastovetsky and Ilya Ledovskih},
title = {Experiments with mpC: Efficient Solving Regular Problems on Heterogeneous Networks of Computers via Irregularization},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
. mpC is a medium-level parallel language for programming heterogeneous networks of computers. It allows to write libraries of parallel routines adaptable to peculiarities of any particular executing multiprocessor system to ensure efficient running. The adaptable routines distribute data and computations in accordance with performances of participating processors. In this case even the problems traditionally considered regular, become irregular. Advantages of mpC for efficient solving of regular problems on heterogeneous networks of computers are demonstrated with an mpC routine implementing Cholesky factorization, with efficiency of the mpC routine being compared with ScaLAPACK one. 1 Introduction A heterogeneous network of computers, being the most common parallel architecture available to common users, can be used for high-performance computing. Taking into account that in the 1990s network capacity increases surpassed processor speed increases [1], pp.6--7, one can predict increa...
Keyphrases
heterogeneous network regular problem mpc routine scalapack one common user efficient solving network capacity increase common parallel architecture high-performance computing processor speed increase medium-level parallel language introduction heterogeneous network parallel routine particular executing multiprocessor system cholesky factorization adaptable routine distribute data
