DMCA
Mesh independent loop fusion for unstructured mesh applications
| Venue: | in Proceedings of the 9th conference on Computing Frontiers, ser. CF ’12 |
| Citations: | 2 - 2 self |
BibTeX
@INPROCEEDINGS{Bertolli_meshindependent,
author = {Carlo Bertolli and Adam Betts and Gihan R. Mudalige and Paul H. J. Kelly and Michael B. Giles},
title = {Mesh independent loop fusion for unstructured mesh applications},
booktitle = {in Proceedings of the 9th conference on Computing Frontiers, ser. CF ’12},
year = {},
pages = {43--52},
publisher = {Online]. Available: http://doi.acm.org/10.1145/2212908.2212917}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Applications based on unstructured meshes are typically compute intensive, leading to long running times. In princi-ple, state-of-the-art hardware, such as multi-core CPUs and many-core GPUs, could be used for their acceleration but these esoteric architectures require specialised knowledge to achieve optimal performance. OP2 is a parallel program-ming layer which attempts to ease this programming burden by allowing programmers to express parallel iterations over elements in the unstructured mesh through an API call, a so-called OP2-loop. The OP2 compiler infrastructure then uses source-to-source transformations to realise a parallel imple-mentation of each OP2-loop and discover opportunities for optimisation. In this paper, we describe how several compiler techniques
Keyphrases
mesh independent loop fusion unstructured mesh application unstructured mesh so-called op2-loop compute intensive api call multi-core cpu state-of-the-art hardware esoteric architecture discover opportunity several compiler technique source-to-source transformation parallel imple-mentation optimal performance op2 compiler infrastructure parallel iteration parallel program-ming layer
