high performance computing on graphics processing units: hgpu.org

Recently, the computational speed and battery capability of mobile devices were greatly prompted. With an enormous number of APPs, users can do many things in mobile devices as well as in computers. Consequently, more and more scientific researchers are encouraged to move their working environment from computers to mobile devices for increasing their work efficiency because they can analyze data and make decisions on their mobile devices anytime and anywhere. Accordingly, we propose a mobile OpenMP programming environment called MOMP in this paper. Using this APP, users can directly write, compile and execute OpenMP programs on their Android-based mobile devices to exploit embedded CPU and GPU for resolving their problems without network connection. Because of source compatibility, MOMP makes users easily port their OpenMP programs from computers to mobile devices without any modification. Moreover, MOMP provides users with an easy interface to choose CPU or GPU for executing different parallel regions in the same program based on the properties of parallel regions. Therefore, MOMP can effectively reduce the programming complexity of heterogeneous computing in mobile devices. Our experimental result also shows that MOMP can effectively exploit the computational power of mobile devices for the performance of user applications.