Parallel programming and performance predictability with Orléans Skeleton Library

摘要

Orléans Skeleton Library (OSL) is a library of parallel algorithmic skeletons in C++ on top of MPI. It provides a structured approach towards parallel programming. Skeletons in OSL are based over the bulk synchronous parallelism model. Applications can be developed using different combinations and compositions of the skeletons. This paper illustrates the expressivity and performance predictability of OSL with two applications: a two dimensional heat diffusion simulation, and an exact N-body simulation. Experiments using these applications are performed on parallel machines.