high performance computing on graphics processing units: hgpu.org

The aim of the project described in this report was to integrate the Molecular Fluid Dynamics simulation package OpenFOAM with the molecular-modelling library OpenMM, in order to enable the execution of Molecular Fluid Dynamics simulations on Graphics Processing Units (GPUs). The reason for this integration, rather than a straightforward substitution of OpenFOAM with OpenMM, was to combine the pre- and post-processing functionality of OpenFOAM (running on CPUs) with fast, GPU-based, Molecular Dynamics implementation provided by OpenMM. The result was a GPUaccelerated Molecular Fluid Dynamics simulation package with a pre- and post-processing capability. Running engineering time- and length-scale Molecular Dynamics simulations entails a considerable computational effort and currently requires computational times that are prohibitively long for transformational advances in fluids engineering to be possible. GPU-accelerated Molecular Fluid Dynamics will enable faster and longer simulations, as well as better exploitation of computational resources. Not only does running simulations on GPUs relieve existing, CPU-based, high-performance computational resources, but it also enables simulations to be run on personal workstations equipped with suitable GPUs, thereby saving on large CPU-cluster waiting times and execution costs. The emphasis of this project was on nano-scale fluid dynamics. The resulting software package will thus be of particular relevance to researchers interested in flows at the smallest scales.