Benchmark Result Run Time Build Time ------------------------------------------------------------- Memory Read 11863 MB/s 1156 ms - Pinned 11863 MB/s 3 ms - Pageable 4908 MB/s 6 ms Memory Write 11425 MB/s 1172 ms - Pinned 11425 MB/s 3 ms - Pageable 4875 MB/s 6 ms Memory Copy 66776 MB/s 4859 ms - 15 MB Block 62241 MB/s 0 ms - 32 MB Block 64489 MB/s 1 ms - 64 MB Block 65019 MB/s 2 ms - 128 MB Block 66072 MB/s 4 ms - 256 MB Block 66574 MB/s 8 ms - 512 MB Block 66667 MB/s 15 ms - 1024 MB Block 66776 MB/s 31 ms - 1024 MB Block 66750 MB/s 31 ms Single-Precision FLOPS 1452 GFLOPS 9016 ms - float1 943 GFLOPS 583 ms 391 ms - float2 1381 GFLOPS 796 ms 313 ms - float4 1447 GFLOPS 760 ms 282 ms - float8 1452 GFLOPS 757 ms 281 ms - float16 1449 GFLOPS 759 ms 281 ms Double-Precision FLOPS 47.13 GFLOPS 9531 ms - double1 47.13 GFLOPS 729 ms 578 ms - double2 46.90 GFLOPS 733 ms 422 ms - double4 46.90 GFLOPS 733 ms 281 ms - double8 46.45 GFLOPS 740 ms 282 ms - double16 46.45 GFLOPS 740 ms 453 ms 24-bit Integer IOPS 496.3 GIOPS 9953 ms - int1 406.7 GIOPS 676 ms 781 ms - int2 496.3 GIOPS 554 ms 343 ms - int4 492.2 GIOPS 558 ms 328 ms - int8 458.0 GIOPS 600 ms 547 ms - int16 404.0 GIOPS 680 ms 1641 ms 32-bit Integer IOPS 495.4 GIOPS 21296 ms - int1 406.6 GIOPS 676 ms 13718 ms - int2 495.4 GIOPS 555 ms 266 ms - int4 494.0 GIOPS 556 ms 250 ms - int8 458.5 GIOPS 600 ms 281 ms - int16 392.8 GIOPS 700 ms 454 ms 64-bit Integer IOPS 83.45 GIOPS 25906 ms - long1 83.45 GIOPS 823 ms 13703 ms - long2 75.16 GIOPS 914 ms 578 ms - long4 66.19 GIOPS 519 ms 672 ms - long8 57.30 GIOPS 600 ms 984 ms - long16 72.17 GIOPS 952 ms 2219 ms AES-256 4006 MB/s 8437 ms - define unroll / 4 MB 3989 MB/s 513 ms 750 ms - define unroll / 8 MB 4006 MB/s 511 ms - define unroll / 16 MB 4000 MB/s 512 ms - define unroll / 32 MB 3995 MB/s 513 ms - define unroll / 64 MB 3977 MB/s 515 ms - define unroll / 128 MB 3943 MB/s 519 ms SHA-1 Hash 12527 MB/s 15265 ms - 512 blocks 4921 MB/s 813 ms 2312 ms - 1024 blocks 9519 MB/s 840 ms - 2048 blocks 12241 MB/s 654 ms - 4096 blocks 11933 MB/s 670 ms - 8192 blocks 11389 MB/s 702 ms - 16384 blocks 11341 MB/s 705 ms - 32768 blocks 11632 MB/s 688 ms - 65536 blocks 12430 MB/s 644 ms - 131072 blocks 12527 MB/s 639 ms Single-Precision Julia 265.6 FPS 8453 ms - float1 break 212.6 FPS 602 ms 1016 ms - float1 stay / unroll 3 265.6 FPS 964 ms 265 ms - float2 stay / unroll 3 239.2 FPS 535 ms 172 ms - float1 stay / unroll 9 241.9 FPS 529 ms 141 ms - float2 stay / unroll 9 215.5 FPS 594 ms 235 ms Double-Precision Mandel 15.58 FPS 6656 ms - double1 break 13.69 FPS 584 ms 156 ms - double1 stay / unroll 3 15.58 FPS 513 ms 156 ms - double2 stay / unroll 3 11.91 FPS 671 ms 312 ms - double1 stay / unroll 9 15.33 FPS 522 ms 156 ms - double2 stay / unroll 9 11.61 FPS 689 ms 172 ms