Task04 Ельцов Данил HSE

[sthfaceless]

Локальный вывод

D:\PycharmProjects\explore\gpu_tasks\cmake-build-debug-mingw\matrix_transpose.exe 1
OpenCL devices:
  Device #0: CPU. 11th Gen Intel(R) Core(TM) i5-11320H @ 3.20GHz. Intel(R) Corporation. Total memory: 16167 Mb
  Device #1: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 8083 Mb
  Device #2: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 7148 Mb
  Device #3: CPU. Microsoft Basic Render Driver. Microsoft. Total memory: 8083 Mb
Using device #1: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 8083 Mb
Data generated for M=4099, K=4109
[matrix_transpose_naive]
    GPU: 0.00223333+-0.00511621 s
    GPU: 7541.55 millions/s
[matrix_transpose_local_bad_banks]
    GPU: 0.00184333+-0.00480508 s
    GPU: 9137.14 millions/s
[matrix_transpose_local_good_banks]
    GPU: 0.00176667+-0.00469988 s
    GPU: 9533.66 millions/s

D:\PycharmProjects\explore\gpu_tasks\cmake-build-debug-mingw\matrix_multiplication.exe 1
OpenCL devices:
  Device #0: CPU. 11th Gen Intel(R) Core(TM) i5-11320H @ 3.20GHz. Intel(R) Corporation. Total memory: 16167 Mb
  Device #1: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 8083 Mb
  Device #2: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 7148 Mb
  Device #3: CPU. Microsoft Basic Render Driver. Microsoft. Total memory: 8083 Mb
Using device #1: GPU. Intel(R) Iris(R) Xe Graphics. Total memory: 8083 Mb
Data generated for M=1025, K=1037, N=1027
CPU: 2.604+-0 s
CPU: 0.768049 GFlops
[naive, ts=4]
    GPU: 0.0484167+-0.00402613 s
    GPU: 41.3081 GFlops
    Average difference: 0%
[naive, ts=8]
    GPU: 0.0469833+-0.000128019 s
    GPU: 42.5683 GFlops
    Average difference: 0%
[naive, ts=16]
    GPU: 0.0469667+-0.000256038 s
    GPU: 42.5834 GFlops
    Average difference: 0%
[local, ts=4]
    GPU: 0.04085+-0.00686737 s
    GPU: 48.9596 GFlops
    Average difference: 0%
[local, ts=8]
    GPU: 0.0144833+-0.00329389 s
    GPU: 138.09 GFlops
    Average difference: 0%
[local, ts=16]
    GPU: 0.0151833+-0.00196207 s
    GPU: 131.723 GFlops
    Average difference: 0.164156%
[local wpt, ts=4, wpt=2]
    GPU: 0.0790333+-0.00267686 s
    GPU: 25.3058 GFlops
    Average difference: 0%
[local wpt, ts=4, wpt=4]
    GPU: 0.1394+-0.00309946 s
    GPU: 14.3472 GFlops
    Average difference: 0%
[local wpt, ts=8, wpt=2]
    GPU: 0.0155667+-0.000495536 s
    GPU: 128.48 GFlops
    Average difference: 0%
[local wpt, ts=8, wpt=4]
    GPU: 0.02575+-0.00715454 s
    GPU: 77.6699 GFlops
    Average difference: 0%
[local wpt, ts=8, wpt=8]
    GPU: 0.0348333+-0.00621602 s
    GPU: 57.4163 GFlops
    Average difference: 0%
[local wpt, ts=16, wpt=2]
    GPU: 0.01185+-0.00651364 s
    GPU: 168.776 GFlops
    Average difference: 0.000303535%
[local wpt, ts=16, wpt=4]
    GPU: 0.0107833+-0.00697876 s
    GPU: 185.471 GFlops
    Average difference: 0%
[local wpt, ts=16, wpt=8]
    GPU: 0.0108+-0.00694934 s
    GPU: 185.185 GFlops
    Average difference: 0%
[local wpt, ts=16, wpt=16]
    GPU: 0.01575+-0.000433013 s
    GPU: 126.984 GFlops
    Average difference: 0%

Вывод Github CI