sharing GPU arrays between processes ( Python )

Accelerated Computing CUDA CUDA Programming and Performance
1

Is there a python3 library to share cudaMalloc-ed memory in GPU among processes? ( I know that it is possible in C++ using CUDA IPC API, but was wondering if there is a way to do it in python.) If not, then how how would you recommend building such library with C++ CUDA IPC code that can be called in python?

Thanks