Highly-Scalable Computational-Based Engineering Algorithms for Emerging Parallel Machine Architectures

ABSTRACT: RNET and The Ohio State University propose to use algorithmic modifications and multi-level parallelization techniques and tools to improve the scalability of the aero-line/aero-elastic coupled CFD/CSD codes relevant to the DoD/AF (e.g., CREATE/Kestrel). The optimizations will address inter-node and intra-node parallelization to better target emerging compute architectures (e.g., multi-cores, GPU, vector units, etc.) and inter-node optimizations to improve scalability to 10's of thousands of cores. In Phase II, the optimizations and modifications identified and prototyped in Phase I will be fully implemented in Kestrel, AVUS, and other relevant Government codes. The CFD optimizations will include algorithmic modifications to the Gauss Seidel solver that improve the cache utilization, data structure modifications to enable vectorization, GPU implementations, and improvements in the compute/communication ratio to improve scalability. In addition, similar inter- and intra-node optimizations will be made to the CSD implementation. Improvements in the CFD/CSD coupling will also be optimized for emerging compute architectures. In Phase III the optimizations will be fully transitioned to the Kestrel, AVUS, and Salinas teams. In addition, other commercial and DoD codes will be optimized using similar techniques. BENEFIT: The scalability advancements will help allow the CREATE/Kestrel project to efficiently reach its goals of providing rapid aeronautic simulations to early phase design and acquisition projects. The scalability improvements will enable the effective usage of the DoD HPC modernization systems for the Kestrel users, allowing large scale simulations to achieve the required rapid turnaround times. In addition, the algorithms, tools, and techniques developed in this project will enable similar scalability improvements in other applications, especially other CFD/CSD applications of interest to the DoD and industry.