MPI Optimizations via MXM and FCA for Maximum Performance on LS-DYNA

From concept to engineering, and from design to test and manufacturing, the automotive industr y relies on powerful virtual development solutions. CFD and crash simulations are performed in an effort to secure quality and accelerate the development process. The recent trends in cluster environments, such as multi-core CPUs, GPUs, cluster file systems and new interconnect speeds and offloading capabilities are changing the dynamics of clustered- based simulations. Software applications are being reshaped for higher parallelism and hardware configuration for solving the new emerging bottlenecks, in order to maintain high scalability and efficiency. In this paper we cover a new co-design architecture with hardware based accelerations and offloads for MPI collectives communications and how it affects LS-DYNA performance.