high performance computing on graphics processing units: hgpu.org

OpenACC is a new specification for a hybrid (CPU + GPU) parallel programming API, in which the programmer uses compiler directives to distribute the computation between the GPU and the CPU. With a similar paradigm to OpenMP, OpenACC presents clear advantages in terms of ease of programming. Regarding performance, however, a comparison between OpenACC and CUDA has not yet been made. This study aims to evaluate OpenACC, establishing a comparison with CUDA. Furthermore, this work aims to identify the main limitations of OpenACC, analyzing its impact on performance. The evaluation is made using three different benchmarks (matrix transpose, dot product and matrix multiplication), each one comprising several implementations. Our results show that, although being in some cases notably slower than optimized CUDA, OpenACC implementations can still benefit from significant performance improvements over serial programs executed on the CPU. Moreover, when compared with less optimized CUDA implementations, OpenACC is shown to provide competitive performance with a much simpler programming model.