In heterogeneous computing, many programming models take use of computation offloading by transfering resource intensive computational tasks to an external platform, such as a cluster, grid or a cloud. Offloading may be necessary due to hardware limitations of a devices, such as limited computational power, storage, and energy. Here, we will make a birdview of the offloading action of CUDA and OpenMP in clang driver.

NDArray is short for N Dimension Array, it is a headers only library for N-Dimensions array(i.e. tensor in deep learning) expressions, such as arithmatic operations, reshape, reduction, slice and etc.