{"created":"2023-07-25T10:24:27.992311+00:00","id":1780,"links":{},"metadata":{"_buckets":{"deposit":"a8d4e240-0344-41d4-800c-6b0b935f9dfe"},"_deposit":{"created_by":1,"id":"1780","owners":[1],"pid":{"revision_id":0,"type":"depid","value":"1780"},"status":"published"},"_oai":{"id":"oai:hiroshima-cu.repo.nii.ac.jp:00001780","sets":["54:383:402"]},"author_link":["10250","10252","10251","10254","10253","10255"],"item_3_biblio_info_7":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2001-03-08","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"22","bibliographicPageEnd":"138","bibliographicPageStart":"133","bibliographicVolumeNumber":"2001","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告. 計算機アーキテクチャ研究会報告"}]}]},"item_3_description_19":{"attribute_name":"フォーマット","attribute_value_mlt":[{"subitem_description":"application/pdf","subitem_description_type":"Other"}]},"item_3_description_5":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"本稿では、ループ運搬依存を含むループ(DOACROSSループ)の並列化に、線形回帰演算の並列化手法を応用する方法について述べる。本手法は、ループのイタレーションの一つ一つを回帰演算の処理単位とみなして線形回帰演算の並列化手法を適用するものである。従来のループ並列化手法とは異なり、ループ全体から並列実行可能な成分を切り分けないため、ループインスタンス内に自然に存在しているデータ局所性が乱されない。アルゴリズム変換により総計算量が増加するものの、粒度は大きく保たれ、計算処理を全てのプロセッサが分担できることから台数効果を得やすい。実験では、細粒度並列処理でしか効果が上がらないと言われる微小ループでも、並列化により数倍程度の高速化が可能であることが確かめられた。 ","subitem_description_type":"Abstract"},{"subitem_description":"In this paper, we show a method of parallel execution of DOACROSS loops utilizing parallel execution techniques for linear recurrences. The method is able to make every processor in a system participate essential computation unlike such methods as normal DOACROSS execution and pipelining, which have critical paths in computation and cannot necessarily utilize all of the processors in a system. A drawback of the method shown in this paper is the increase of the amount of computation in order to cut serial chains of dependences. Experimental results show the method is feasible enough and efficient on both shared-memory systems and distributed-machine environments.","subitem_description_type":"Abstract"}]},"item_3_publisher_8":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会"}]},"item_3_relation_12":{"attribute_name":"論文ID（NAID）","attribute_value_mlt":[{"subitem_relation_type":"isIdenticalTo","subitem_relation_type_id":{"subitem_relation_type_id_text":"110004029064","subitem_relation_type_select":"NAID"}}]},"item_3_relation_17":{"attribute_name":"関連サイト","attribute_value_mlt":[{"subitem_relation_name":[{"subitem_relation_name_text":"http://www.ipsj.or.jp/"}],"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://www.ipsj.or.jp/","subitem_relation_type_select":"URI"}}]},"item_3_rights_15":{"attribute_name":"権利","attribute_value_mlt":[{"subitem_rights":"ここに掲載した著作物の利用に関する注意：本著作物の著作権は（社）情報処理学会に帰属します。本著作物は著作権者である情報処理学会の許可のもとに掲載するものです。ご利用に当たっては「著作権法」ならびに「情報処理学会倫理綱領」に従うことをお願いいたします。 "},{"subitem_rights":"The copyright of this material is retained by the Information Processing Society of Japan (IPSJ). This material is published on this web site with the agreement of the author (s) and the IPSJ. Please be complied with Copyright Law of Japan and the Code of Ethics of the IPSJ if any users wish to reproduce, make derivative work, distribute or make available to the public any part or whole thereof. All Rights Reserved, Copyright (C) Information Processing Society of Japan."}]},"item_3_source_id_11":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10096105","subitem_source_identifier_type":"NCID"}]},"item_3_version_type_20":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_version_resource":"http://purl.org/coar/version/c_970fb48d4fbd8a85","subitem_version_type":"VoR"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"川端, 英之"},{"creatorName":"カワバタ, ヒデユキ","creatorNameLang":"ja-Kana"}],"nameIdentifiers":[{"nameIdentifier":"10250","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"谷口, 宏美"},{"creatorName":"タニグチ, ヒロミ","creatorNameLang":"ja-Kana"}],"nameIdentifiers":[{"nameIdentifier":"10251","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"津田, 孝夫"},{"creatorName":"ツダ, タカオ","creatorNameLang":"ja-Kana"}],"nameIdentifiers":[{"nameIdentifier":"10252","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"KAWABATA, Hideyuki","creatorNameLang":"en"}],"nameIdentifiers":[{"nameIdentifier":"10253","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"TANIGUCHI, Hiromi","creatorNameLang":"en"}],"nameIdentifiers":[{"nameIdentifier":"10254","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"TSUDA, Takao","creatorNameLang":"en"}],"nameIdentifiers":[{"nameIdentifier":"10255","nameIdentifierScheme":"WEKO"}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2023-05-26"}],"displaytype":"detail","filename":"IPSJ-ARC00142023.pdf","filesize":[{"value":"716.9 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"IPSJ-ARC00142023.pdf","url":"https://hiroshima-cu.repo.nii.ac.jp/record/1780/files/IPSJ-ARC00142023.pdf"},"version_id":"f11bdf3f-0410-4e20-a56e-ead15e8e41e5"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"technical report","resourceuri":"http://purl.org/coar/resource_type/c_18gh"}]},"item_title":"線形回帰演算の並列化手法の応用によるDOACROSSループの並列実行","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"線形回帰演算の並列化手法の応用によるDOACROSSループの並列実行"},{"subitem_title":"Bulk Recurrent Parallel Execution of DOACROSS Loops","subitem_title_language":"en"}]},"item_type_id":"3","owner":"1","path":["402"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-05-26"},"publish_date":"2023-05-26","publish_status":"0","recid":"1780","relation_version_is_last":true,"title":["線形回帰演算の並列化手法の応用によるDOACROSSループの並列実行"],"weko_creator_id":"1","weko_shared_id":1},"updated":"2023-07-25T10:38:27.655812+00:00"}