|Data Locality Optimization Strategies for AMR Applications on GPU-accelerated Supercomputers
Mohamed Wahib (RIKEN Advanced Institute for Computational Science)
|Adaptive Mesh Refinement methods reduce computa- tional requirements of problems by increasing resolution for only areas of interest. However, in practice, efficient AMR implementations are difficult considering that the mesh hierarchy management must be optimized for the underlying hardware. Architecture complexity of GPUs can render efficient AMR to be particularity challenging in GPU-accelerated supercomputers. This poster presents a high-level framework that can automatically transform serial uniform mesh code annotated by the user into par- allel adaptive mesh code optimized for GPU-accelerated clusters. We show experimental results on three produc- tion applications. The speedups of code generated by our framework are comparable to hand-written AMR code while achieving good and weak scaling up to 1000 GPUs.
|研究紹介ポスター ／ 最終報告書