OpenACC is a high-level programming model for accelerators, such as NVIDIA GPUs, that allows programmers to accelerate applications using compiler directives to specify loops and regions of code in standard C, C++ and Fortran to be offloaded to an accelerator. Through the use of compiler directives, OpenACC allows programmers to maintain a single source code for the CPU and GPU that is portable��
]]>