Research Article
Accelerating the SCE-UA Global Optimization Method Based on Multi-Core CPU and Many-Core GPU
Table 3
Total execution time (seconds) of the CUDA-SCE-UA.
| nopt | ngs | 4 | 8 | 16 | 32 | 64 | 128 | 256 | 512 | 1024 |
| 10 | 224.62 | 228.95 | 241.62 | 251.60 | 245.62 | 247.45 | 260.40 | 264.61 | 267.10 | 20 | 406.99 | 412.23 | 409.77 | 412.14 | 414.13 | 396.97 | 416.59 | 420.15 | 424.20 | 30 | 670.82 | 648.42 | 604.67 | 626.22 | 629.26 | 611.52 | 619.27 | 623.42 | 624.14 | 40 | 980.73 | 937.47 | 886.81 | 860.83 | 805.71 | 814.37 | 814.33 | 818.35 | 839.48 | 50 | 1400.51 | 1173.31 | 1081.71 | 1078.16 | 1115.32 | 1046.00 | 1080.26 | 1053.03 | 1064.78 |
|
|