cuda-samples/bin/x86_64/linux/release/APM_UnifiedMemoryPerf.txt
2023-03-01 01:41:29 +00:00

18 lines
860 B
Plaintext

GPU Device 0: "Hopper" with compute capability 9.0
Running ........................................................
Overall Time For matrixMultiplyPerf
Printing Average of 20 measurements in (ms)
Size_KB UMhint UMhntAs UMeasy 0Copy MemCopy CpAsync CpHpglk CpPglAs
4 0.210 0.264 0.332 0.014 0.033 0.026 0.037 0.024
16 0.201 0.307 0.489 0.025 0.043 0.035 0.046 0.045
64 0.311 0.381 0.758 0.067 0.084 0.075 0.074 0.063
256 0.545 0.604 1.429 0.323 0.228 0.212 0.197 0.187
1024 1.551 1.444 2.436 1.902 0.831 0.784 0.714 0.728
4096 4.960 4.375 7.863 11.966 3.239 3.179 2.908 2.919
16384 18.911 17.022 29.696 77.375 13.796 13.757 12.874 12.862
NOTE: The CUDA Samples are not meant for performance measurements. Results may vary when GPU Boost is enabled.