GPU-STREAM Version: 0.0 Implementation: CUDA Precision: double Warning: array size must divide 1024 Resizing array from 50000000 to 49999872 Array size: 381.5 MB (=0.4 GB) Total size: 1144.4 MB (=1.1 GB) Using CUDA device Tesla K40c Function MBytes/sec Min (sec) Max Average Copy 220178.580 0.00363 0.00364 0.00364 Mul 219898.608 0.00364 0.00364 0.00364 Add 220694.367 0.00544 0.00544 0.00544 Triad 220662.874 0.00544 0.00545 0.00544