Jetson Nano or Lowest End GT720 notebook for CUDA learning parallel programming.

Accelerated Computing CUDA CUDA Programming and Performance
1

Greetings,

I am a newbie and I want to learn CUDA parallel programming in GPU.

I want to set up a working bench to start learning it. I come to these 2 choices. I have either jetson or a lower-end notebook with supported CUDA GPU.

In the future, I want to make a demo from the basic algorithm (such as palindrome) until let say generating pi number using GPU.

And other things, to achieve that is C the only language that supported now? Is it possible to do it in python?

Thoughts or inputs are mostly welcome.

thank you in advance.

2

well there is PyCuda and Numba, both of which allow you to write Kernels inline in Python code.

In pyCuda the kernels have to be written in CUDA C. Numba seems to have its own dialect.

For other languages that can map on the instruction set architecture of CUDA, have a look at
https://developer.nvidia.com/cuda-fortran