wgblas
    Preparing search index...

    Module benchmarks/nvidia-geforce-gtx-1650/snrm2

    Benchmark results for snrm2 on Nvidia Geforce Gtx 1650.

    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    32 0.0164 0.0078 0.0182 0.0070 111.6%
    64 0.0164 0.0156 0.0180 0.0142 110.0%
    128 0.0162 0.0316 0.0180 0.0284 111.1%
    512 0.0164 0.1250 0.0180 0.1139 109.7%
    1024 0.0161 0.2545 0.0180 0.2270 112.1%
    4096 0.0164 1.0000 0.0180 0.9094 110.0%
    16384 0.0166 3.9537 0.0184 3.5586 111.1%
    65536 0.0180 14.6025 0.0181 14.4480 101.1%
    262144 0.0288 36.4697 0.0190 55.0723 66.2%
    1048576 0.0636 65.8984 0.0442 94.9797 69.4%
    4194304 0.2105 79.6851 0.1436 116.8590 68.2%
    16777216 0.9362 71.6816 2.7987 23.9783 298.9%

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

    snrm2-default GB/s chart

    snrm2-default ms chart

    • snrm2.js — WebGPU benchmark script
    • snrm2.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 0.0217 0.0059 0.0181 0.0071 83.2%
    64 0.0213 0.0120 0.0180 0.0142 84.6%
    128 0.0212 0.0242 0.1469 0.0035 690.5%
    512 0.0210 0.0974 0.0180 0.1139 85.5%
    1024 0.0210 0.1950 0.0191 0.2140 91.1%
    4096 0.0212 0.7711 0.0180 0.9127 84.5%
    16384 0.0222 2.9553 0.0180 3.6441 81.1%
    65536 0.0287 9.1429 0.0188 13.9557 65.5%
    262144 0.0477 21.9772 0.0372 28.1875 78.0%
    1048576 0.1153 36.3685 0.1052 39.8819 91.2%
    4194304 0.3909 42.9181 0.3767 44.5369 96.4%

    snrm2-stride4 GB/s chart

    snrm2-stride4 ms chart

    Nvidia Geforce Gtx 1650 — stride = 32
    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    32 0.0220 0.0058 0.1618 0.0008 725.7%
    64 0.0214 0.0120 0.0194 0.0132 90.8%
    128 0.0215 0.0238 0.0182 0.0281 84.7%
    512 0.0210 0.0974 0.0175 0.1173 83.0%
    1024 0.0212 0.1936 0.0173 0.2370 81.7%
    4096 0.0219 0.7491 0.0177 0.9259 80.9%
    16384 0.0299 2.1927 0.0396 1.6529 132.7%
    65536 0.0463 5.6575 0.1685 1.5555 363.7%
    262144 0.1145 9.1556 0.2785 3.7647 243.2%
    1048576 0.3914 10.7173 0.7664 5.4724 195.8%

    snrm2-stride32 GB/s chart

    snrm2-stride32 ms chart

    Nvidia Geforce Gtx 1650 — stride = 256
    n wgblas ms wgblas GB/s cuBLAS ms cuBLAS GB/s efficiency
    512 0.0274 0.0748 0.0176 0.1162 64.3%
    1024 0.0279 0.1469 0.0181 0.2265 64.8%
    4096 0.0284 0.5763 0.0181 0.9030 63.8%
    16384 0.0385 1.7003 0.0490 1.3381 127.1%
    65536 0.0650 4.0305 0.0719 3.6474 110.5%

    snrm2-stride256 GB/s chart

    snrm2-stride256 ms chart

    See also:

    Settings
    Member Visibility

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