Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
井戸村 泰宏; 仲田 資季; 山田 進; 町田 昌彦; 今村 俊幸*; 渡邉 智彦*; 沼波 政倫*; 井上 晃*; 堤 重信*; 三吉 郁夫*; et al.
International Journal of High Performance Computing Applications, 28(1), p.73 - 86, 2014/02
被引用回数：17 パーセンタイル：76.23(Computer Science, Hardware & Architecture)A plasma turbulence research based on five dimensional (5D) gyrokinetic simulations is one of the most critical and demanding issues in fusion science. To pioneer new physics regimes both in problem sizes and in time scales, an improvement of strong scaling is essential. In this work, the strong scaling and the parallel efficiency of a gyrokinetic toroidal 5D Eularian code GT5D is dramatically improved by novel developing communication overlap techniques. On the K-computer, excellent strong scaling is achieved beyond 100 k cores with keeping the sustained performance of 
10% ( 307 TFlops using 196,608 cores), and simulations for next generation large-scale fusion experiments are significantly accelerated.
井戸村 泰宏; 仲田 資季; 山田 進; 町田 昌彦; 今村 俊幸*; 渡邉 智彦*; 沼波 政倫*; 井上 晃*; 堤 重信*; 三吉 郁夫*; et al.
Proceedings of 31st JSST Annual Conference; International Conference on Simulation Technology (JSST 2012) (USB Flash Drive), p.234 - 242, 2012/09
A plasma turbulence research based on 5D gyrokinetic simulations is one of the most critical and demanding issues in fusion science. To pioneer new physics regimes both in problem sizes and in time scales, an improvement of strong scaling is essential. Overlap of computations and communications is a promising approach to improve strong scaling, but it often fails on practical applications with conventional MPI libraries. In this work, this classical issue is clarified, and resolved by developing communication overlap techniques with mpi_test and communication threads, which work even on conventional MPI libraries and network hardwares. These techniques dramatically improve the parallel efficiency of a gyrokinetic Eularian code GT5D on K and Helios, which adopt dedicated and commodity networks.
井戸村 泰宏; 仲田 資季; 山田 進; 町田 昌彦; 今村 俊幸*; 渡邉 智彦*; 沼波 政倫*; 井上 晃*; 堤 重信*; 三吉 郁夫*; et al.
no journal, ,
A plasma turbulence research based on 5D gyrokinetic simulations is one of the most critical and demanding issues in fusion science. To pioneer new physics regimes both in problem sizes and in time scales, an improvement of strong scaling is essential. Overlap of computations and communications is a promising approach to improve strong scaling, but it often fails on practical applications with conventional MPI libraries. In this work, this classical issue is clarified, and resolved by developing communication overlap techniques with mpi_test and communication threads. These techniques dramatically improve the parallel efficiency of a gyrokinetic Eularian code GT5D on the K-computer, and ITER-size simulations are significantly accelerated.
井戸村 泰宏; 仲田 資季; 山田 進; 町田 昌彦; 今村 俊幸*; 渡邉 智彦*; 沼波 政倫*; 井上 晃*; 堤 重信*; 三吉 郁夫*; et al.
no journal, ,
5次元ジャイロ運動論モデルに基づくプラズマ乱流シミュレーションは、核融合プラズマにおける乱流輸送現象を解析するための標準的な手法となっている。国際熱核融合実験炉ITERのような将来の大型装置を解析するうえで、シミュレーションの適用範囲、特に、装置サイズの規模を拡大することが重要となっており、より大きな計算機資源が必要とされている。しかしながら、現在のペタスケール計算機、あるいは、将来のエクサスケール計算機の能力を引き出すには、10万コアを超える並列度というハードウェアの厳しい要求を満たす必要がある。本研究では、差分法に基づくプラズマ乱流コードGT5D[Idomura, CPC2009]において、新たな通信マスク手法の開発を行い、京において約99.9998%という従来に比べて一桁以上高い並列化率を達成し、約20万コアまで良好なストロングスケーリングを確認した。
