Notes on GPU Programming with Julia

I’m planning to use Julia as the language in my intro to parallel computing course next year. My personal favorite type of parallel programming is on GPUs, so I’ve started with some notes on GPU programming with Julia, which may be of interest to others.


I don’t know anything about GPU programming, but… your linked document seems much more complex than GPU programming with Python, like on these posts: 1,2,3 or 4

Is there any easy/practical tutorial for GPU programming in Julia for those that know nothing about CS ?

Yes, I do assume a CS background. Have you looked at the CUDA.jl introduction? I’d suggest starting with CuArrays. They key is to have enough work to keep many GPU threads busy, and to use Julia’s dot notation. A good way to learn is to explore on your own.