Solving molecular distance geometry problems in OpenCL

We focus on the following computational chemistry problem: Given a subset of the exact distances between atoms, reconstruct the three-dimensional position of each atom in the given molecule. The distance matrix is generally sparse. This problem is both important and challenging. Our contribution is a novel combination of two known techniques (parallel breadth-first search and geometric buildup) and its OpenCL parallel implementation. The approach has the potential to speed up computation of three-dimensional structures of molecules - a critical process in computational chemistry. From experiments on multi-core CPUs and graphic processing units, we conclude that, for sufficient large problems, our implementation shows a moderate scalability