ダウンロード数: 382
このアイテムのファイル:
ファイル | 記述 | サイズ | フォーマット | |
---|---|---|---|---|
ipsjjip.22.642.pdf | 1.62 MB | Adobe PDF | 見る/開く |
タイトル: | Parallel Hierarchical Matrices with Adaptive Cross Approximation on Symmetric Multiprocessing Clusters |
著者: | Ida, Akihiro Iwashita, Takeshi ![]() ![]() Mifune, Takeshi ![]() ![]() Takahashi, Yasuhito |
著者名の別形: | 伊田, 明弘 |
キーワード: | boundary element method matrix approximation hierarchical matrices adaptive cross approximation parallel scalability symmetric multiprocessing clusters |
発行日: | Oct-2014 |
出版者: | Information Processing Society of Japan |
誌名: | Journal of Information Processing |
巻: | 22 |
号: | 4 |
開始ページ: | 642 |
終了ページ: | 650 |
抄録: | We discuss a scheme for hierarchical matrices with adaptive cross approximation on symmetric multiprocessing clusters. We propose a set of parallel algorithms that are applicable to hierarchical matrices. The proposed algorithms are implemented using the flat-MPIand hybrid MPI+OpenMP programming models. The performance of these implementations is evaluated using an electric field analysis computed on two symmetric multiprocessing cluster systems. Although the flat-MPI version gives better parallel scalability when constructing hierarchical matrices, the speed-up reaches a limit in the hierarchical matrix-vector multiplication. We succeeded in developing a hybrid MPI+OpenMP version to improve the parallel scalability. In numerical experiments, the hybrid version exhibits a better parallel speed-up for the hierarchical matrix-vector multiplication up to 256 cores. |
著作権等: | © 2014 by the Information Processing Society of Japan |
URI: | http://hdl.handle.net/2433/191244 |
DOI(出版社版): | 10.2197/ipsjjip.22.642 |
出現コレクション: | 学術雑誌掲載論文等 |
![](/dspace/image/articlelinker.gif)
このリポジトリに保管されているアイテムはすべて著作権により保護されています。