wgblas
    Preparing search index...

    Module benchmarks/nvidia-geforce-gtx-1650/sswap

    Benchmark results for sswap on Nvidia Geforce Gtx 1650.

    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    32 0.0060 0.0858 0.0039 0.1311 65.4%
    64 0.0059 0.1739 0.0039 0.2645 65.8%
    128 0.0060 0.3422 0.0036 0.5689 60.2%
    512 0.0059 1.3875 0.0041 2.0000 69.4%
    1024 0.0060 2.7380 0.0040 4.1124 66.6%
    4096 0.0061 10.6667 0.0038 17.3559 61.5%
    16384 0.0064 41.1658 0.0041 63.5039 64.8%
    65536 0.0081 129.0079 0.0082 128.0000 100.8%
    262144 0.0284 147.9368 0.0287 146.2857 101.1%
    1048576 0.1019 164.6892 0.1024 163.8400 100.5%
    4194304 0.3953 169.7824 0.4035 166.3350 102.1%
    16777216 1.5624 171.8114 1.6353 164.1462 104.7%

    Efficiency = wgblas GB/s ÷ cuBLAS GB/s × 100. 100% means parity with cuBLAS; values above 100% mean wgblas achieved greater throughput.

    sswap-default GB/s chart

    sswap-default ms chart

    • sswap.js — WebGPU benchmark script
    • sswap.c — CUDA / cuBLAS reference script

    Unless noted otherwise, every result above uses unit stride (incx = incy = 1) — the normal case, and the coalesced, best-case GPU access pattern. Real usage sometimes passes a non-unit stride (e.g. operating on a row or column of a larger matrix, where incx = lda), which breaks memory coalescing and costs measurably more. This section sweeps a few representative strides to characterize that cost separately, collapsed below by default — expand a stride to see its table and chart.

    Nvidia Geforce Gtx 1650 — stride = 4
    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    32 0.0060 0.0856 0.0030 0.1693 50.5%
    64 0.0061 0.1684 0.0031 0.3333 50.5%
    128 0.0059 0.3459 0.0032 0.6465 53.5%
    512 0.0061 1.3474 0.0035 2.3167 58.2%
    1024 0.0061 2.6667 0.0033 4.9709 53.6%
    4096 0.0067 9.7757 0.0036 18.2044 53.7%
    16384 0.0083 31.5077 0.0054 48.9075 64.4%
    65536 0.0273 38.3700 0.0269 39.0095 98.4%
    262144 0.1014 41.3803 0.1004 41.7959 99.0%
    1048576 0.3957 42.3993 0.3907 42.9463 98.7%
    4194304 1.5692 42.7659 1.5633 42.9287 99.6%

    sswap-stride4 GB/s chart

    sswap-stride4 ms chart

    Nvidia Geforce Gtx 1650 — stride = 32
    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    32 0.0061 0.0836 0.0040 0.1285 65.0%
    64 0.0061 0.1667 0.0041 0.2500 66.7%
    128 0.0061 0.3351 0.0039 0.5246 63.9%
    512 0.0063 1.3061 0.0048 1.7239 75.8%
    1024 0.0065 2.5160 0.0050 3.2508 77.4%
    4096 0.0092 7.1483 0.0063 10.4224 68.6%
    16384 0.0351 7.4642 0.0342 7.6740 97.3%
    65536 0.1371 7.6507 0.1331 7.8769 97.1%
    262144 0.5448 7.6992 0.5262 7.9711 96.6%
    1048576 2.1811 7.6921 2.9476 5.6918 135.1%

    sswap-stride32 GB/s chart

    sswap-stride32 ms chart

    Nvidia Geforce Gtx 1650 — stride = 256
    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    512 0.0063 1.2962 0.0054 1.5193 85.3%
    1024 0.0066 2.4976 0.0051 3.2000 78.0%
    4096 0.0105 6.2630 0.0091 7.1860 87.2%
    16384 0.0643 4.0756 0.0712 3.6826 110.7%
    65536 0.2925 3.5853 0.2892 3.6264 98.9%

    sswap-stride256 GB/s chart

    sswap-stride256 ms chart

    See also:

    Settings
    Member Visibility

    On This Page
    Nvidia Geforce Gtx 1650 — wgblas vs cuBLASSee alsoStride sweep