2016
DOI: 10.1007/s11227-016-1696-9
|View full text |Cite
|
Sign up to set email alerts
|

Applying vectorization of diagonal sparse matrix to accelerate wind field calculation

Abstract: Wind field calculation is a critical issue in reaching accurate forest fire propagation predictions. However, when the involved terrain map is large, the amount of memory and the execution time can prevent them from being useful in an operational environment. Wind field calculation involves sparse matrices that are usually stored in CSR storage format. This storage format can cause sparse matrix-vector multiplications to create a bottleneck due to the number of cache misses involved. Moreover, the matrices inv… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(1 citation statement)
references
References 5 publications
0
1
0
Order By: Relevance
“…Therefore, computation time has become a principal constraint if the whole system is to deliver useful information to end-users in due time. Recent efforts to parallelize WindNinja exploit domain decomposition methods Sanjuan et al (2016c), computational parallelization based on GPU (graphics processing unit) (Sanjuan et al, 2016a) and hybrid integration using message passing interface (MPI) and open multiprocessing (OpenMP) (Sanjuan et al, 2016b). Despite the remarkable speed-up achieved with those strategies, the computing time required for a single run of a 1500×1500-cell map exceeds 90 s on a 64-node processor using the most efficient approach (hybrid MPI-OpenMP integration).…”
Section: Introductionmentioning
confidence: 99%
“…Therefore, computation time has become a principal constraint if the whole system is to deliver useful information to end-users in due time. Recent efforts to parallelize WindNinja exploit domain decomposition methods Sanjuan et al (2016c), computational parallelization based on GPU (graphics processing unit) (Sanjuan et al, 2016a) and hybrid integration using message passing interface (MPI) and open multiprocessing (OpenMP) (Sanjuan et al, 2016b). Despite the remarkable speed-up achieved with those strategies, the computing time required for a single run of a 1500×1500-cell map exceeds 90 s on a 64-node processor using the most efficient approach (hybrid MPI-OpenMP integration).…”
Section: Introductionmentioning
confidence: 99%