Geodesic
Performance study of the architecture-independent VGL framework for efficient implementation of graph algorithms
Numerical methods and programming, Tome 24 (2023) no. 4, pp. 485-499.

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

Graph algorithms are currently often used to solve various modeling tasks, since many real-life objects are well modeled by graphs (for example, a road network or social connections). At the same time, the efficient implementation of such algorithms is often very complex, which is due, in particular, to irregular memory access when working with graphs and the huge size of the input graphs. Graph frameworks — software environments for implementing graph algorithms — can help solve this problem. Previously, an architecture-independent VGL (Vector Graph Library) framework was developed that allows for efficient implementation of graph algorithms on various hardware platforms (multi-core processors with vector extensions, graphics accelerators and NEC vector processors). In this work, the performance of VGL was studied on different platforms, its performance was also compared with existing analogues, and an approach for automatic selection of input graph format based on machine learning methods was proposed and evaluated.
Keywords: graph framework; graph algorithms; high-performance computing; performance analysis; vector processing; VGL. 
@article{VMP_2023_24_4_a10,
     author = {D. I. Lichmanov and I. V. Afanasyev and Vl. V. Voevodin},
     title = {Performance study of the architecture-independent {VGL} framework for efficient implementation of graph algorithms},
     journal = {Numerical methods and programming},
     pages = {485--499},
     publisher = {mathdoc},
     volume = {24},
     number = {4},
     year = {2023},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2023_24_4_a10/}
}
TY  - JOUR
AU  - D. I. Lichmanov
AU  - I. V. Afanasyev
AU  - Vl. V. Voevodin
TI  - Performance study of the architecture-independent VGL framework for efficient implementation of graph algorithms
JO  - Numerical methods and programming
PY  - 2023
SP  - 485
EP  - 499
VL  - 24
IS  - 4
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2023_24_4_a10/
LA  - ru
ID  - VMP_2023_24_4_a10
ER  - 
%0 Journal Article
%A D. I. Lichmanov
%A I. V. Afanasyev
%A Vl. V. Voevodin
%T Performance study of the architecture-independent VGL framework for efficient implementation of graph algorithms
%J Numerical methods and programming
%D 2023
%P 485-499
%V 24
%N 4
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2023_24_4_a10/
%G ru
%F VMP_2023_24_4_a10
D. I. Lichmanov; I. V. Afanasyev; Vl. V. Voevodin. Performance study of the architecture-independent VGL framework for efficient implementation of graph algorithms. Numerical methods and programming, Tome 24 (2023) no. 4, pp. 485-499. http://geodesic.mathdoc.fr/item/VMP_2023_24_4_a10/