Conference paperKUNOVSKÝ Jiří, KRAUS Michal, ŠIMEK Václav and PETŘEK Jiří. GPU Based Accelleration of Telegraph equation. In: Proceedings UKSim 10th International Conference EUROSIM/UKSim2008. Cambridge: IEEE Computer Society, 2008, pp. 629630. ISBN 0769531148.  Publication language:  english 

Original title:  GPU Based Accelleration of Telegraph equation 

Title (cs):  Akcelerace výpočtu telegrafní rovnice založená na GPU 

Pages:  629630 

Proceedings:  Proceedings UKSim 10th International Conference EUROSIM/UKSim2008 

Conference:  10th International Conference on Computer Modelling and Simulation 

Place:  Cambridge, GB 

Year:  2008 

ISBN:  0769531148 

Publisher:  IEEE Computer Society 

Keywords 

Taylor series, GPU, accelleration, parallel algorithm, telegraph equation

Annotation 

In a matter of just a few years, the programmable graphics processor unit has evolved into an absolute computing workhorse. With multiple cores driven by very high memory bandwidth, today's GPUs offer incredible resources for both graphics and nongraphics processing. An original mathematical method "Modern Taylor Series Method" (MTSM) which uses the Taylor series method for solving differential equations in a nontraditional way has been developed and implemented in TKSL software. Even though this method is not much preferred in the literature, experimental calculations have shown and theoretical analyses have veriffed that the accuracy and stability of the Taylor series method exceeds the currently used algorithms for numerically solving differential equations. It is the aim of the paper to illustrate GPU and MTSM for numerical solutions of a telegraph line. 
BibTeX: 

@INPROCEEDINGS{
author = {Ji{\v{r}}{\'{i}} Kunovsk{\'{y}} and Michal Kraus and
V{\'{a}}clav {\v{S}}imek and Ji{\v{r}}{\'{i}} Pet{\v{r}}ek},
title = {GPU Based Accelleration of Telegraph equation},
pages = {629630},
booktitle = {Proceedings UKSim 10th International Conference
EUROSIM/UKSim2008},
year = {2008},
location = {Cambridge, GB},
publisher = {IEEE Computer Society},
ISBN = {0769531148},
language = {english},
url = {http://www.fit.vutbr.cz/research/view_pub.php.en.iso88592?id=8596}
} 
