An Improved Distance Matrix Computation Algorithm for Multicore Clusters

Table 6

DistVect1 comparison against other programs.

Software	Platform	Maximum sequence length	Highest GCUPS	Highest speedup
CUDASW++ 2	Dual-GPU GeForce GTX 295	59 k	28 (144–5478)	1.78 with (CUDASW++ 1.0)
CUDASW++ 3	Dual-GPU GeForce GTX 690	59 k	185.60 (144–5478)	3.2 with (CUDASW++ 2.0)
MC64-ClustalWP2	Intel Xeon Quad Core	300 k	0.92	7.76 with (ClustalW-MPI)
SSE2	Cell/BE	0.858 k	0.0058	76.20 with ClustalW sequential
ClustalW-MPI	PC cluster	1100	0.07	14.5 with ClustalW sequential
DistVect1	Cluster of multicores	163 k	11.69	9.2 with (ClustalW-MPI)