Benchmark Result Run Time Build Time ------------------------------------------------------------- Memory Read 1595 MB/s 1782 ms - Pinned 1595 MB/s 19 ms - Pageable 1288 MB/s 23 ms Memory Write 1448 MB/s 1453 ms - Pinned 1448 MB/s 21 ms - Pageable 807 MB/s 37 ms Memory Copy 12189 MB/s 4938 ms - 15 MB Block 11213 MB/s 3 ms - 32 MB Block 11996 MB/s 5 ms - 64 MB Block 12088 MB/s 11 ms - 128 MB Block 12135 MB/s 21 ms - 256 MB Block 12157 MB/s 42 ms - 512 MB Block 12160 MB/s 84 ms - 512 MB Block 12189 MB/s 84 ms Single-Precision FLOPS 624.7 GFLOPS 18594 ms - float1 483.8 GFLOPS 568 ms 7359 ms - float2 617.0 GFLOPS 891 ms 531 ms - float4 624.7 GFLOPS 880 ms 391 ms - float8 563.2 GFLOPS 976 ms 360 ms - float16 583.2 GFLOPS 943 ms 375 ms Double-Precision FLOPS 30.51 GFLOPS 8860 ms - double1 30.46 GFLOPS 564 ms 453 ms - double2 30.51 GFLOPS 563 ms 500 ms - double4 30.51 GFLOPS 563 ms 469 ms - double8 30.27 GFLOPS 568 ms 328 ms - double16 30.02 GFLOPS 572 ms 359 ms 24-bit Integer IOPS 122.0 GIOPS 8891 ms - int1 122.0 GIOPS 563 ms 265 ms - int2 122.0 GIOPS 563 ms 328 ms - int4 122.0 GIOPS 563 ms 453 ms - int8 122.0 GIOPS 563 ms 484 ms - int16 121.1 GIOPS 568 ms 937 ms 32-bit Integer IOPS 122.0 GIOPS 18859 ms - int1 122.0 GIOPS 563 ms 1547 ms - int2 122.0 GIOPS 563 ms 219 ms - int4 122.0 GIOPS 563 ms 250 ms - int8 122.0 GIOPS 563 ms 390 ms - int16 121.1 GIOPS 568 ms 515 ms 64-bit Integer IOPS 30.16 GIOPS 20391 ms - long1 25.98 GIOPS 661 ms 1407 ms - long2 26.49 GIOPS 649 ms 359 ms - long4 29.36 GIOPS 585 ms 422 ms - long8 30.16 GIOPS 570 ms 593 ms - long16 28.94 GIOPS 594 ms 938 ms AES-256 1214 MB/s 13203 ms - define loop / 4 MB 1213 MB/s 844 ms 313 ms - define loop / 8 MB 1214 MB/s 844 ms - define loop / 16 MB 1213 MB/s 844 ms - define loop / 32 MB 1209 MB/s 847 ms - define loop / 64 MB 1202 MB/s 852 ms - define loop / 128 MB 1188 MB/s 862 ms SHA-1 Hash 2660 MB/s 15875 ms - 512 blocks 1497 MB/s 668 ms 734 ms - 1024 blocks 2270 MB/s 881 ms - 2048 blocks 2660 MB/s 752 ms - 4096 blocks 1915 MB/s 522 ms - 8192 blocks 1917 MB/s 522 ms - 16384 blocks 1872 MB/s 534 ms - 32768 blocks 1897 MB/s 527 ms - 65536 blocks 1863 MB/s 1074 ms Single-Precision Julia 124.7 FPS 12203 ms - float1 break 121.9 FPS 525 ms 297 ms - float1 stay / unroll 3 116.7 FPS 548 ms 281 ms - float2 stay / unroll 3 122.2 FPS 524 ms 407 ms - float4 stay / unroll 3 113.1 FPS 566 ms 219 ms - float1 stay / unroll 9 124.7 FPS 513 ms 218 ms - float2 stay / unroll 9 112.8 FPS 567 ms 203 ms Double-Precision Mandel 9.43 FPS 8859 ms - double1 break 8.65 FPS 925 ms 282 ms - double1 stay / unroll 3 9.25 FPS 865 ms 266 ms - double2 stay / unroll 3 7.14 FPS 560 ms 266 ms - double1 stay / unroll 9 9.43 FPS 849 ms 297 ms - double2 stay / unroll 9 7.16 FPS 558 ms 297 ms