Industrial-Quality Porting of the Army's Helios CFD Code into Xeon-Phi-Enabled Systems - A Process-Based Approach for Maximal Speedup

The recent introduction of Many-Integrated-Core (MIC) architectures such as the Intel Xeon Phi presents an opportunity for enormous performance increases in numerically intensive HPC codes. This project aims at capitalizing on this opportunity by substantially accelerating DoDs CFD solvers by means of maximal utilization of the Xeon Phis resources. In addition to providing increased performance on a DoD code, the proposed work will produce a library of fundamental kernels for exploiting the power of Xeon Phi on CFD applications. The resulting technology will be accompanied by a carefully-planned technology transfer strategy that will maximize the impact of the resulting solvers and tools into Armys strategic missions. The proposed work will leverage Accelogics extensive experience gathered from the acceleration of both Cartesian and unstructured CFD solvers for US Government Agencies. In Phase I we will develop a proof-of-concept prototype that will demonstrate the technologys ability to provide DoDs target acceleration. In Phase II we will work on maturing and refining the technology, driven by a concrete target of accelerating a DoDs High Performance CFD framework. We have secured complementary funds in the amount of $100,000 to ensure that the proposed Phase I work will be successfully accomplished in a timely manner.