Programming in CUDA for Kepler and Maxwell Architecture

Since the first version of CUDA was launch, many improvements were made in GPU computing. Every new CUDA version included important novel features, turning this architecture more and more closely related to a typical parallel High Performance Language. This tutorial will present the GPU architecture and CUDA principles, trying to conceptualize novel features included by NVIDIA, such as dynamics parallelism, unified memory and concurrent kernels. This text also includes some optimization remarks for CUDA programs. 1Instituto de Computação, Universidade Federal Fluminense, Niterói RJ, Brazil {[email protected]} 2Depto. de Ciência da Computação, Universidade Federal Rural do Rio de Janeiro, Nova Iguaçu RJ, Brazil {[email protected]} Programming in CUDA for Kepler and Maxwell Architecture