Limits...
CBESW: sequence alignment on the Playstation 3.

Wirawan A, Kwoh CK, Hieu NT, Schmidt B - BMC Bioinformatics (2008)

Bottom Line: As a result, the computational power needed by bioinformatics applications is growing exponentially as well.For large datasets, our implementation on the PlayStation 3 provides a significant improvement in running time compared to other implementations such as SSEARCH, Striped Smith-Waterman and CUDA.The results from our experiments demonstrate that the PlayStation 3 console can be used as an efficient low cost computational platform for high performance sequence alignment applications.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Computer Engineering, Nanyang Technological University, Singapore. adri0004@ntu.edu.sg

ABSTRACT

Background: The exponential growth of available biological data has caused bioinformatics to be rapidly moving towards a data-intensive, computational science. As a result, the computational power needed by bioinformatics applications is growing exponentially as well. The recent emergence of accelerator technologies has made it possible to achieve an excellent improvement in execution time for many bioinformatics applications, compared to current general-purpose platforms. In this paper, we demonstrate how the PlayStation 3, powered by the Cell Broadband Engine, can be used as a computational platform to accelerate the Smith-Waterman algorithm.

Results: For large datasets, our implementation on the PlayStation 3 provides a significant improvement in running time compared to other implementations such as SSEARCH, Striped Smith-Waterman and CUDA. Our implementation achieves a peak performance of up to 3,646 MCUPS.

Conclusion: The results from our experiments demonstrate that the PlayStation 3 console can be used as an efficient low cost computational platform for high performance sequence alignment applications.

Show MeSH
Performance comparison with the CUDA implementation on a single Nvidia GeForce 8800GTX. Performance comparison between our CBESW implementation with CUDA implementation on a single Nvidia GeForce 8800GTX, in terms of MCUPS. All queries were run against Swiss-Prot release 55.2. Seventeen query sequences with lengths of 63 to 852 amino acids were used. The scoring matrix used for the CUDA implementation was BLOSUM 50.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2571991&req=5

Figure 8: Performance comparison with the CUDA implementation on a single Nvidia GeForce 8800GTX. Performance comparison between our CBESW implementation with CUDA implementation on a single Nvidia GeForce 8800GTX, in terms of MCUPS. All queries were run against Swiss-Prot release 55.2. Seventeen query sequences with lengths of 63 to 852 amino acids were used. The scoring matrix used for the CUDA implementation was BLOSUM 50.

Mentions: The performance comparison between the PS3 implementation and CUDA-SW on one Nvidia GeForce 8800GTX is shown in Figure 8. The CUDA implementation experiment was conducted with a GeForce 8800GTX 512 MB installed in a PC with a Dual-Core AMD Opteron 2210 1.8 GHz CPU, 2 GB RAM running Fedora 6. The substitution matrix used is BLOSUM50. As can be seen from the figure, our implementation achieves a better MCUPS performance. The PS3 peak performance is 3 times faster compared to the peak performance CUDA implementation on a single Nvidia GeForce 8800GTX.


CBESW: sequence alignment on the Playstation 3.

Wirawan A, Kwoh CK, Hieu NT, Schmidt B - BMC Bioinformatics (2008)

Performance comparison with the CUDA implementation on a single Nvidia GeForce 8800GTX. Performance comparison between our CBESW implementation with CUDA implementation on a single Nvidia GeForce 8800GTX, in terms of MCUPS. All queries were run against Swiss-Prot release 55.2. Seventeen query sequences with lengths of 63 to 852 amino acids were used. The scoring matrix used for the CUDA implementation was BLOSUM 50.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2571991&req=5

Figure 8: Performance comparison with the CUDA implementation on a single Nvidia GeForce 8800GTX. Performance comparison between our CBESW implementation with CUDA implementation on a single Nvidia GeForce 8800GTX, in terms of MCUPS. All queries were run against Swiss-Prot release 55.2. Seventeen query sequences with lengths of 63 to 852 amino acids were used. The scoring matrix used for the CUDA implementation was BLOSUM 50.
Mentions: The performance comparison between the PS3 implementation and CUDA-SW on one Nvidia GeForce 8800GTX is shown in Figure 8. The CUDA implementation experiment was conducted with a GeForce 8800GTX 512 MB installed in a PC with a Dual-Core AMD Opteron 2210 1.8 GHz CPU, 2 GB RAM running Fedora 6. The substitution matrix used is BLOSUM50. As can be seen from the figure, our implementation achieves a better MCUPS performance. The PS3 peak performance is 3 times faster compared to the peak performance CUDA implementation on a single Nvidia GeForce 8800GTX.

Bottom Line: As a result, the computational power needed by bioinformatics applications is growing exponentially as well.For large datasets, our implementation on the PlayStation 3 provides a significant improvement in running time compared to other implementations such as SSEARCH, Striped Smith-Waterman and CUDA.The results from our experiments demonstrate that the PlayStation 3 console can be used as an efficient low cost computational platform for high performance sequence alignment applications.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Computer Engineering, Nanyang Technological University, Singapore. adri0004@ntu.edu.sg

ABSTRACT

Background: The exponential growth of available biological data has caused bioinformatics to be rapidly moving towards a data-intensive, computational science. As a result, the computational power needed by bioinformatics applications is growing exponentially as well. The recent emergence of accelerator technologies has made it possible to achieve an excellent improvement in execution time for many bioinformatics applications, compared to current general-purpose platforms. In this paper, we demonstrate how the PlayStation 3, powered by the Cell Broadband Engine, can be used as a computational platform to accelerate the Smith-Waterman algorithm.

Results: For large datasets, our implementation on the PlayStation 3 provides a significant improvement in running time compared to other implementations such as SSEARCH, Striped Smith-Waterman and CUDA. Our implementation achieves a peak performance of up to 3,646 MCUPS.

Conclusion: The results from our experiments demonstrate that the PlayStation 3 console can be used as an efficient low cost computational platform for high performance sequence alignment applications.

Show MeSH