Computer Science – Research and Development manuscript No. (will be inserted by the editor) Application of the ParalleX Execution Model to Stencil-based Problems

Abstract In the prospect of the upcoming exa-scale era with millions of execution units, the question of how to deal with this level of parallelism efficiently is of time-critical relevance. State-of-the-Art parallelization techniques such as OpenMP and MPI are not guaran-teed to solve the expected problems of starvation, grow-ing latencies, overheads, and contention. On the other hand, new parallelization paradigms promise to effi-ciently hide latencies and contain starvation and con-tention. In this paper we analyze the performance of one novel parallelization strategy for shared and distributed memory machines. We will focus on shared memory ar-chitectures and compare the performance of the Par-alleX execution model against the quasi-standard Open-MP for a standard stencil-based problem. We compare in detail the OpenMP implementation of two applica-tions of Jacobi solvers (one based on regular grid and one based on an irregular grid structure) with the cor-responding implementation of these applications using HPX (High Performance ParalleX), the first feature-complete, open-source implementation of ParalleX, an