Skip to content

This project demonstrates the integration of a CUDA kernel within an NVIDIA Holoscan application. It consists of two custom operators: one for memory allocation and data initialization, and another for executing the CUDA kernel. The application was profiled using Nsight systems and the kernel with Nsight compute

Notifications You must be signed in to change notification settings

abhiMishra98/Holoscan-Add-Matrices

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

Holoscan Add Matrices

This project demonstrates a simple matrix addition application using NVIDIA Holoscan with CUDA.


Build Instructions

  1. Create a build folder and navigate inside:
mkdir build
cd build
  1. Configure the project using CMake (make sure to set the Holoscan path):
cmake ../src -DCMAKE_PREFIX_PATH=/opt/nvidia/holoscan
  1. Build the project:
make -j

Profiling using Nsight systems

Once, makefile is generated run the following code

nsys profile   --trace=cuda,nvtx,osrt   --gpu-metrics-devices=all   --sample=cpu   --output=profile_report3   ./cudaTest

Open Nsight systems UI and locate the file

nsys-ui

Profiling using Nsight compute

Once, makefile is generated run the following code

ncu --replay-mode application -o compute_report ./cudaTest

Open Nsight compute directly with the file name as an argument

ncu-ui compute_report.ncu-rep

About

This project demonstrates the integration of a CUDA kernel within an NVIDIA Holoscan application. It consists of two custom operators: one for memory allocation and data initialization, and another for executing the CUDA kernel. The application was profiled using Nsight systems and the kernel with Nsight compute

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published