{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00018422","sets":["934:1119:1136:1140"]},"path":["1140"],"owner":"1","recid":"18422","title":["ソフトウェア分散共有メモリを用いたマクロデータフロー処理"],"pubdate":{"attribute_name":"公開日","attribute_value":"2005-03-15"},"_buckets":{"deposit":"3088bd26-a29c-475f-8585-a31898508929"},"_deposit":{"id":"18422","pid":{"type":"depid","value":"18422","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"ソフトウェア分散共有メモリを用いたマクロデータフロー処理","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"ソフトウェア分散共有メモリを用いたマクロデータフロー処理"},{"subitem_title":"Macro-data-flow Using Software Distributed Shared Memory","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"並列処理","subitem_subject_scheme":"Other"}]},"item_type_id":"3","publish_date":"2005-03-15","item_3_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"電気通信大学大学院情報システム学研究科"},{"subitem_text_value":"電気通信大学大学院情報システム学研究科"},{"subitem_text_value":"電気通信大学大学院情報システム学研究科"}]},"item_3_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Systems The University of Electro-Communications","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Systems The University of Electro-Communications","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Systems The University of Electro-Communications","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/18422/files/IPSJ-TACS4604007.pdf"},"date":[{"dateType":"Available","dateValue":"2007-03-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-TACS4604007.pdf","filesize":[{"value":"1.1 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"16"},{"tax":["include_tax"],"price":"0","billingrole":"11"},{"tax":["include_tax"],"price":"0","billingrole":"14"},{"tax":["include_tax"],"price":"0","billingrole":"15"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"081500c1-166b-4d68-8051-fc5af5521174","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2005 by the Information Processing Society of Japan"}]},"item_3_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"田邊, 浩志"},{"creatorName":"本多, 弘樹"},{"creatorName":"弓場, 敏嗣"}],"nameIdentifiers":[{}]}]},"item_3_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hiroshi, Tanabe","creatorNameLang":"en"},{"creatorName":"Hiroki, Honda","creatorNameLang":"en"},{"creatorName":"Toshitsugu, Yuba","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_3_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11833852","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_3_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7829","subitem_source_identifier_type":"ISSN"}]},"item_3_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"ループ並列処理の限界を超えてさらなる性能向上のために，ループやサブルーチンの粗粒度タスク（マクロタスク）レベルの並列性を利用するマクロデータフロー処理が注目されている．マクロデータフロー処理を分散メモリシステム上で実現するためには，異なるプロセッサに割り当てられたマクロタスク間でデータ授受を行う機能が必要となる．これに対し，我々は明示的な通信によってデータを授受するデータ到達条件による実行方式を提案しているが，コンパイル時にデータ参照を正確に解析できないプログラムでは不要なデータ転送をしてしまい，性能低下が問題となる．本稿では，マクロタスク間のデータ授受にソフトウェア分散共有メモリを用い，必要に応じたデータ転送を行う方式を提案する．2 つのページベースソフトウェア分散共有メモリのTreadMarks とJIAJIA を用いてPC クラスタ上で提案方式の実装と性能評価をした．その結果，不規則なデータ参照のプログラムに対して不要なデータ転送を削減でき，データ到達条件による実行方式に比べて最大25%の性能向上が得られた．","subitem_description_type":"Other"}]},"item_3_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Macro-dataflow processing, which exploits a parallelism among coarse grain tasks (macrotasks) such as loops and subroutines, is considered promising to break the performance limits of loop parallelism. To realize macro-dataflow processing on distributed memory systems,“data reaching conditions”, a method to make a sender-receiver pair of a data transfer determined at runtime, has been proposed. However, irregular data accesses induce extra data transfers, which lead to performance deteriorations. This paper proposes an implementation scheme using software distributed shared memory, which enables on-demand data fetching. This paper describes the implementations using two well-accepted page-based Software Distributed Shared Memory systems, TreadMarks and JIAJIA. Evaluation results on a PC cluster show software distributed memory approach is up to 25% faster than the “data reaching conditions”.","subitem_description_type":"Other"}]},"item_3_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"68","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌コンピューティングシステム（ACS）"}],"bibliographicPageStart":"56","bibliographicIssueDates":{"bibliographicIssueDate":"2005-03-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"SIG4(ACS9)","bibliographicVolumeNumber":"46"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"id":18422,"updated":"2025-01-22T22:46:27.854962+00:00","links":{},"created":"2025-01-18T22:51:09.854079+00:00"}