WEKO3
アイテム
Pipelined task control using OpenMP
https://ipsj.ixsq.nii.ac.jp/records/233171
https://ipsj.ixsq.nii.ac.jp/records/2331712293012d-5ce6-4454-b046-98aa1274a99a
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
IPSJ-HPC24193029.pdf (1.9 MB)
2026年3月11日からダウンロード可能です。
|
Copyright (c) 2024 by the Information Processing Society of Japan
|
|
| 非会員：¥660, IPSJ:学会員：¥330, HPC:会員：¥0, DLIB:会員：¥0 | ||
| Item type | SIG Technical Reports(1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2024-03-11 | |||||||||
| タイトル | ||||||||||
| タイトル | Pipelined task control using OpenMP | |||||||||
| タイトル | ||||||||||
| 言語 | en | |||||||||
| タイトル | Pipelined task control using OpenMP | |||||||||
| 言語 | ||||||||||
| 言語 | eng | |||||||||
| キーワード | ||||||||||
| 主題Scheme | Other | |||||||||
| 主題 | 自動チューニング・OpenMP | |||||||||
| 資源タイプ | ||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_18gh | |||||||||
| 資源タイプ | technical report | |||||||||
| 著者所属 | ||||||||||
| Department of Electrical Engineering and Information System, The University of Tokyo | ||||||||||
| 著者所属 | ||||||||||
| Information Technology Center, The University of Tokyo／Department of Electrical Engineering and Information System, The University of Tokyo | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Department of Electrical Engineering and Information System, The University of Tokyo | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Information Technology Center, The University of Tokyo / Department of Electrical Engineering and Information System, The University of Tokyo | ||||||||||
| 著者名 |
Tianya, Wu
× Tianya, Wu
× Toshihiro, Hanawa
|
|||||||||
| 著者名(英) |
Tianya, Wu
× Tianya, Wu
× Toshihiro, Hanawa
|
|||||||||
| 論文抄録 | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | In recent High-Performance Computing (HPC) systems, in order to achieve higher performance, the number of cores in a compute node is increasing. However, there are “unused” cores during a certain phase, such as file IO for snapshots in typical computational science applications with in-situ visualization. In this study, to flexibly control parallelism in such codes, we studied to realize pipelined structure using OpenMP. Our method, applied to GOTHIC, an N-body calculation code, reduced processing time by 30% compared to the previous version and 55% compared to the original. | |||||||||
| 論文抄録(英) | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | In recent High-Performance Computing (HPC) systems, in order to achieve higher performance, the number of cores in a compute node is increasing. However, there are “unused” cores during a certain phase, such as file IO for snapshots in typical computational science applications with in-situ visualization. In this study, to flexibly control parallelism in such codes, we studied to realize pipelined structure using OpenMP. Our method, applied to GOTHIC, an N-body calculation code, reduced processing time by 30% compared to the previous version and 55% compared to the original. | |||||||||
| 書誌レコードID | ||||||||||
| 収録物識別子タイプ | NCID | |||||||||
| 収録物識別子 | AN10463942 | |||||||||
| 書誌情報 |
研究報告ハイパフォーマンスコンピューティング（HPC） 巻 2024-HPC-193, 号 29, p. 1-7, 発行日 2024-03-11 |
|||||||||
| ISSN | ||||||||||
| 収録物識別子タイプ | ISSN | |||||||||
| 収録物識別子 | 2188-8841 | |||||||||
| Notice | ||||||||||
| SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. | ||||||||||
| 出版者 | ||||||||||
| 言語 | ja | |||||||||
| 出版者 | 情報処理学会 | |||||||||
