Shared-Memory Parallelization of the Fast Marching Method Using an Overlapping Domain-Decomposition Approach

摘要

The fast marching method is used to compute a monotone front propagation of anisotropic nature by solving the eikonal equation. Due to the sequential nature of the original algorithm, parallel approaches presented so far were unconvincing. In this work, we introduce a shared-memory parallelization approach which is based on an overlapping domain decomposition technique. We introduce our parallel algorithm of the fast marching method tailored to shared-memory environments and discuss benchmark results based on a C++ implementation using OpenMP. We compare the sequential execution performance as well as the accuracy with reference implementations of the fast marching method and the fast iterative method; the latter is also used to evaluate the parallel scalability. Our shared-memory parallel fast marching method convinces both with regard to serial and parallel execution performance as well as with respect to accuracy.