深入理解 Nsight System 与 Nsight Compute 性能分析优化工具

Helps understand the profiler’s output in app’s algorithmic context

Call the API functions from your source

Link the NVTX library on the compiler command line with –lnvToolsExt

CuPy -> https://docs.cupy.dev/en/v9.0.0a1/reference/cuda.html#profiler

TF -> https://developer.nvidia.com/blog/tensorflow-performance-logging-plugin-nvtx-plugins-