Title :
Parallel multiple sequences alignment in SMP cluster
Author :
Tan, Guangming ; Feng, Shengzhong ; Sun, Ninghui
Author_Institution :
Inst. of Comput. Technol., Chinese Acad. of Sci., Beijing
Abstract :
Multiple sequences alignment is a fundamental and challenging problem in computational molecular biology. It is commonly used to analyse the DNA/protein sequences. To develop a high efficient parallel algorithm is a very important solution to speedup this time consuming problem. However, due to the irregular computation behaviors based on tree, it is difficult to achieve good load balancing, so the utilization of processor is very low. This paper presents a parallel multiple sequences alignment algorithm featuring a mixed fine and coarse grained parallelization approach. The parallel algorithm is suitable to be implemented in SMP cluster, which is the main architecture of current cluster systems. We implemented the parallel algorithm in SMP cluster using a hybrid MPI/OpenMP method and the experimental results shows that the mixed fine and coarse algorithm achieves higher speedup
Keywords :
biology computing; message passing; molecular biophysics; open systems; parallel algorithms; resource allocation; shared memory systems; workstation clusters; DNA sequence analysis; SMP cluster; computational molecular biology; hybrid MPI/OpenMP; load balancing; parallel algorithm; parallel multiple sequences alignment; protein sequence analysis; Approximation algorithms; Biology computing; Clustering algorithms; Concurrent computing; DNA; Parallel algorithms; Polynomials; Proteins; Sequences; Sun;
Conference_Titel :
High-Performance Computing in Asia-Pacific Region, 2005. Proceedings. Eighth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7695-2486-9
DOI :
10.1109/HPCASIA.2005.70