Table 8 Running times (milliseconds) of CUDA implementation
From: Solving the maximum subsequence sum and related problems using BSP/CGM model and multi-GPU CUDA
n | CS-1 4GPUs | CS-1 2GPUs | CS-1 1GPU | CS-3 | CS-5 | CS-2 |
|---|---|---|---|---|---|---|
220 | 9.033 | 10.338 | 4.040 | 4.701 | 19.220 | 2.650 |
221 | 15.794 | 12.300 | 5.194 | 7.839 | 36.470 | 4.310 |
222 | 15.661 | 14.138 | 7.456 | 14.254 | 71.342 | 7.662 |
223 | 15.053 | 16.320 | 11.999 | 27.1798 | 139.110 | 14.505 |
224 | 16.036 | 17.480 | 20.913 | 53.415 | 304.978 | 28.348 |
225 | 23.632 | 31.100 | 39.653 | – | 573.287 | 56.360 |
226 | 30.764 | 56.451 | 76.051 | – | – | – |
227 | 55.126 | 92.363 | 149.896 | – | – | – |
228 | 103.174 | 178.439 | – | – | – | – |
229 | 204.072 | – | – | – | – | – |