OpenACC is an API which allows parallel programmers to provide simple hints, known as “directives,” to the compiler, identifying which areas of code to accelerate, without requiring programmers to modify or adapt the underlying code itself. By exposing parallelism to the compiler, directives allow the compiler to do the detailed work of mapping the computation onto the accelerator

What problem is OpenACC solving?

o It provides a simple programming model for accelerators that can work in the presence or absence of an accelerator

o It defines a unified set of directives for programming accelerators in portable standards-compliant C, C++ or Fortran

· What are the benefits of OpenACC?

o Enables an easy path for developers to start reaping the benefits from powerful many-core accelerators like GPUs.

o Permits a common code base for accelerated and non-accelerator enabled systems. OpenACC directives can be ignored on non-accelerator enabled systems.

o The specification allows for potential implementations on different types and brands of accelerators (CPUs, GPUs)

o Incremental path for moving legacy applications to accelerators, that may disturb the existing code less than other approaches.

o Allows programmer tools to focus on supporting a common accelerator standard with their own unique extensions.