cuda-lab

Playing with CUDA and GPUs in Google Colab.

Usage

1. Open a Colab notebook: https://colab.research.google.com/
2. Create a new Python 3 notebook
3. Change runtime type selecting GPU as hardware accelerator
4. Git clone this repository:

!git clone https://github.com/alessandrobessi/cuda-lab.git

5. Change permissions:

!chmod 755 cuda-lab/INSTALL.sh

6. Install cuda, nvcc, gcc, and g++:

!./cuda-lab/INSTALL.sh

7. Add /usr/local/cuda/bin to PATH:

import os
os.environ['PATH'] += ':/usr/local/cuda/bin'

8. Compile an existing Cuda source:

!nvcc cuda-lab/add.cu -o add -Wno-deprecated-gpu-targets

9. Run the compiled Cuda source using the Nvidia profiler tool:

!nvprof ./add

10. or just time it:

!time ./add

You can also create a Cuda source file using the magic command %%writefile <filename.cu>:

%%writefile snippet.cu
#include <stdlib.h>
#include <stdio.h>
#include <math.h>
...

and then compile and run it!

!nvcc snippet.cu -o snippet -Wno-deprecated-gpu-targets
!nvprof ./snippet