CUDA-Sankoff: Using GPU to Accelerate the Pairwise Structural RNA Alignment
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
CUDA-Sankoff : Using GPU to Accelerate the Pairwise Structural RNA Alignment. / Sundfeld, Daniel; Havgaard, Jakob H.; Gorodkin, Jan; De Melo, Alba C.M.A.
Proceedings: 2017 25th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, PDP 2017. IEEE, 2017. p. 295-302 7912663.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - CUDA-Sankoff
T2 - 25th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, PDP 2017
AU - Sundfeld, Daniel
AU - Havgaard, Jakob H.
AU - Gorodkin, Jan
AU - De Melo, Alba C.M.A.
PY - 2017
Y1 - 2017
N2 - In this paper, we propose and evaluate CUDASankoff, a solution to the RNA structural alignment problem based on the Sankoff algorithm in Graphics Processing Units (GPUS). To our knowledge, this is the first time the Sankoff algorithm is implemented in GPU. In our solution, we show how to linearize the Sankoff 4-dimensional dynamic programming (4D DP) matrix and we propose a two-level wavefront approach to exploit the parallelism. The results were obtained with two different NVidia GPUS, comparing sets of real RNA sequences with lengths from 46 to 281 nucleotides. We show that our GPU approach is up to 24 times faster than a 16-core CPU solution in the 281 nucleotide Sankoff execution.
AB - In this paper, we propose and evaluate CUDASankoff, a solution to the RNA structural alignment problem based on the Sankoff algorithm in Graphics Processing Units (GPUS). To our knowledge, this is the first time the Sankoff algorithm is implemented in GPU. In our solution, we show how to linearize the Sankoff 4-dimensional dynamic programming (4D DP) matrix and we propose a two-level wavefront approach to exploit the parallelism. The results were obtained with two different NVidia GPUS, comparing sets of real RNA sequences with lengths from 46 to 281 nucleotides. We show that our GPU approach is up to 24 times faster than a 16-core CPU solution in the 281 nucleotide Sankoff execution.
KW - CUDA
KW - GPU
KW - RNA
KW - Sankoff Algorithm
U2 - 10.1109/PDP.2017.15
DO - 10.1109/PDP.2017.15
M3 - Article in proceedings
AN - SCOPUS:85019629536
SP - 295
EP - 302
BT - Proceedings
PB - IEEE
Y2 - 6 March 2017 through 8 March 2017
ER -
ID: 184388992