Geodesic
A parallel program model for execution time estimation
Numerical methods and programming, Tome 23 (2022) no. 1, pp. 13-28.

Voir la notice de l'article provenant de la source Math-Net.Ru

Programs for general-purpose graphics processing units represented as kernels without indefinite loops are considered in this paper. Such kernels can be implemented by CUDA or OpenCL technologies, for example. For execution time estimation, various models of program execution are introduced: from very “naive” to more reliable. All models are presented in the form of matrix expressions in max-plus algebra.
Keywords: parallel programs, CUDA, OpenCL
Mots-clés : max-plus algebra. 
@article{VMP_2022_23_1_a4,
     author = {V. A. Antonyuk and N. G. Mikheev},
     title = {A parallel program model for execution time estimation},
     journal = {Numerical methods and programming},
     pages = {13--28},
     publisher = {mathdoc},
     volume = {23},
     number = {1},
     year = {2022},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2022_23_1_a4/}
}
TY  - JOUR
AU  - V. A. Antonyuk
AU  - N. G. Mikheev
TI  - A parallel program model for execution time estimation
JO  - Numerical methods and programming
PY  - 2022
SP  - 13
EP  - 28
VL  - 23
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2022_23_1_a4/
LA  - ru
ID  - VMP_2022_23_1_a4
ER  - 
%0 Journal Article
%A V. A. Antonyuk
%A N. G. Mikheev
%T A parallel program model for execution time estimation
%J Numerical methods and programming
%D 2022
%P 13-28
%V 23
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2022_23_1_a4/
%G ru
%F VMP_2022_23_1_a4
V. A. Antonyuk; N. G. Mikheev. A parallel program model for execution time estimation. Numerical methods and programming, Tome 23 (2022) no. 1, pp. 13-28. http://geodesic.mathdoc.fr/item/VMP_2022_23_1_a4/