Using the PyCUDA module
PyCUDA is a Python wrap for Compute Unified Device Architecture (CUDA), the software library developed by NVIDIA for GPU programming. The CUDA programming model is the starting point of understanding how to program the GPU properly with PyCUDA. There are concepts that must be understood and assimilated to be able to approach this tool correctly and to understand the more specific topics that are covered in the following recipes.
A hybrid programming model
The programming model "hybrid" of CUDA (and consequently of PyCUDA, which is a Python wrapper) is implemented through specific extensions to the standard library of the C language. These extensions have been created, whenever possible, syntactically like the function calls in the standard C library. This allows a relatively simple approach to a hybrid programming model that includes the host and device code. The management of the two logical parts is done by the NVCC compiler. Here is a brief description...
