研究者詳細 - 高橋　徹

講演・口頭発表等 - 高橋　徹

分割表示 37 件中 1 - 20 件目／全件表示 >>

境界要素法の高速化

高橋徹

日本学術会議　第４回計算力学シンポジウム

境界要素法とレベルセット法を用いた2次元電磁波動問題における多目的トポロジー最適化

北林達也, 飯盛浩司, 高橋徹, 松本敏郎

日本機械学会第27回計算力学講演会（CMD2014）

格子ボルツマン法による3次元流れ場のトポロジー最適化

原彰宏, 松本敏郎, 山田崇恭, 高橋徹, 飯盛浩司

日本機械学会第27回計算力学講演会（CMD2014）

GPU統合型CPUを用いた二次元ヘルムホルツ方程式に対する境界要素法の高速化

大澤尚希, 高橋徹, 吉田翔太, 飯盛浩司, 松本敏郎

日本機械学会第27回計算力学講演会（CMD2014）

波動問題のBurton-Miller型積分方程式に関するアイソジオメトリック境界要素法の開発

榛葉祐太, 高橋徹, 林慎一, 坂部遥輝, 飯盛浩司, 松本敏郎

日本機械学会第27回計算力学講演会（CMD2014）

２次元境界要素法に関する高速多重極型直接解法の実装と性能評価

高橋徹, 飯盛浩司, 松本敏郎

日本機械学会第27回計算力学講演会（CMD2014）

高速直接境界要素法を用いた2次元電磁波動問題のトポロジー感度解析について

飯盛浩司, 高橋徹, 松本敏郎

日本機械学会第27回計算力学講演会（CMD2014）

An interpolation-based fast multipole boundary integral equation method for wave equation 国際会議

Toru Takahashi

12th U.S. National Congress on Computational Mechanics (USNCCM12)

開催年月日： 2013年7月

記述言語：英語会議種別：口頭発表（一般）

国名：アメリカ合衆国

From the mathematical and numerical viewpoint, the enhancement of the fast multipole method (FMM; Greengard et al, 1987) to time-dependent problems is an interesting and challenging topic in the research field of the boundary integral equation method (BIEM). However, as far as I know, there have been only several works regarding the FMM in time domain, e.g., Strain 1994, Ergin et al 1998, Shanker et al 2003, Takahashi et al 2003, and Tausch 2007. Following these studies, this study purposes to develop a new fast time-domain boundary integral equation method (TDBIEM) for wave equation in three dimensions. The present method is based on the interpolation of the time-dependent Green's function to re-express the function in the form of separation of variables. This enables to construct a time-domain variant of the FMM. The computational complexity of the resulting TDBIEM is estimated to O(N^1.33*M), while that of the conventional TDBIEM is O(N^2*M), where N and M denote the spatial and temporal degrees of freedom, respectively. In this talk, I will mention the theoretical and numerical aspects of the fast method. Acknowledgement: This work was supported by JSPS KAKENHI Grant Number 24656072.

A time-domain BIEM for Wave Equation accelereted by Fast Multipole Method using Interpolation 国際会議

Toru Takahashi

The 11th International Conference on Mathematical and Numerical Aspects of Waves (Waves 2013)

Task-based Parallelization of the Fast Multipole Method on NVIDIA GPUs and Multicore Processors 国際会議

Emmanuel Agullo, Bérenger Bramas, Olivier Coulaud, Matthias Messner, Eric Darve, Toru Takahashi

GPU Technology Conference (GTC) 2013

開催年月日： 2013年3月

記述言語：英語会議種別：口頭発表（一般）

国名：アメリカ合衆国

Learn about the fast multipole method (FMM) and its optimization on NVIDIA GPUs. The FMM is a well-known algorithm with a variety of applications in areas like galaxy simulation, electrostatic potential calculations, boundary element methods, integral equations, dislocations dynamics, etc. The FMM offers several difficulties when running on parallel heterogeneous platforms such as multicore processors with GPUs. Some parts of the calculation suffer from limited concurrency, and load-balancing can be very uneven for certain distributions of particles. We will present a new API and runtime system, called StarPU, that allows expressing a calculation as a graph of tasks, with dependencies, and contains a runtime system that can optimally schedule those tasks on a parallel machine. StarPU supports conventional multicore processors as well as NVIDIA GPUs.

An Interpolation-based Fast Multipole Boundary Integral Equation Method for Wave Equation

Toru Takahashi

第62回理論応用力学講演会(NCTAM2013)

Pipelining the Fast Multipole Method over a Runtime System 国際会議

Emmanuel Agullo, Berenger Bramas, Olivier Coulaud, Eric F. Darve, Matthias Messner, Toru Takahashi

SIAM Conference on Computational Science and Engineering

開催年月日： 2013年2月 - 2013年3月

記述言語：英語会議種別：口頭発表（一般）

国名：アメリカ合衆国

Fast multipole methods (FMM) usually require a careful
tuning of the algorithm for both the targeted physics and
the hardware. In this talk, we propose a new approach
that achieves high performance across architectures. We
express the FMM algorithm as a task flow and employ a
runtime system to process the tasks on the different processing
units. We carefully design the task flow, the mathematical
operators, their CPU and GPU implementations
and their schedule.

An FMM for Isogeometric BEM for Potential Problems in 2D

Toru Takahashi, Yuya Nakai, Toshiro Matsumoto

JSME 25th Computational Mechanics Conference

大規模音響問題における高速多重極アイソジオメトリック境界要素法の開発

林慎一，高橋徹，松本敏郎

日本機械学会計算力学講演会講演論文集

An acceleration of multipole-to-local operation for low-frequency fast multipole BIEM with GPU 国際会議

Toru Takahashi, Cris Cecka, Eric Darve

Symposium of the International Association for Boundary Element Methods (IABEM) 2011

An implementation of multipole-to-local (M2L) operation for low-frequency fast multipole BIEM on GPU 国際会議

Toru Takahashi, Cris Cecka, Eric Darve

11th US National Congress on Computational Mechanics

Helmholtz方程式に対する低周波高速多重極境界積分方程式法のGPUへの実装

Toru Takahashi, Cris Cecka, Eric Darve

日本機械学会第23回計算力学講演会

An acceleration of the black-box fast multipole method with GPU

2nd International Workshops on Advances in Computational Mechanics

境界要素法におけるGPUの活用例

高橋徹

第17回　計算数理工学フォーラム

The black-box fast multipole method on graphical processing units 国際会議

10th US National Congress on Computational Mechanics

　前のページ - 次のページ　

<高橋 徹>

講演・口頭発表等 - 高橋 徹

<高橋　徹>

講演・口頭発表等 - 高橋　徹