{"id":182841,"updated":"2025-01-20T03:56:14.212467+00:00","links":{},"created":"2025-01-19T00:50:24.842370+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00182841","sets":["1164:1579:9047:9217"]},"path":["9217"],"owner":"11","recid":"182841","title":["実行フェーズを考慮したトランザクショナルメモリのスケジューリング手法"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-07-19"},"_buckets":{"deposit":"9a2d7a97-ca90-4f92-a4e3-21735fa95a03"},"_deposit":{"id":"182841","pid":{"type":"depid","value":"182841","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"実行フェーズを考慮したトランザクショナルメモリのスケジューリング手法","author_link":["399437","399433","399434","399435","399436"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"実行フェーズを考慮したトランザクショナルメモリのスケジューリング手法"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"自動並列化・コード最適化","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2017-07-19","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"名古屋工業大学"},{"subitem_text_value":"名古屋工業大学"},{"subitem_text_value":"名古屋大学"},{"subitem_text_value":"国立情報学研究所"},{"subitem_text_value":"名古屋工業大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Nagoya Institute of Technology","subitem_text_language":"en"},{"subitem_text_value":"Nagoya Institute of Technology","subitem_text_language":"en"},{"subitem_text_value":"Nagoya University","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Informatics","subitem_text_language":"en"},{"subitem_text_value":"Nagoya Institute of Technology","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/182841/files/IPSJ-ARC17227040.pdf","label":"IPSJ-ARC17227040.pdf"},"date":[{"dateType":"Available","dateValue":"2019-07-19"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-ARC17227040.pdf","filesize":[{"value":"528.6 kB"}],"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":"44"}],"accessrole":"open_date","version_id":"77feed91-029f-4b15-806a-26f78155ba0d","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2017 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"多治見, 知紀"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"廣田, 杏珠"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"塩谷, 亮太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"五島, 正裕"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"津邑, 公暁"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10096105","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8574","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"マルチコア環境では，一般的にロックを用いて共有変数へのアクセスを調停する．しかし，ロックにはデッドロックの発生や並列度の低下などの問題があるため，ロックを使用しない並行性制御機構としてトランザクショナルメモリ （TM） が提案されている．この機構をハードウェア上に実装したハードウェアトランザクショナルメモリ （HTM） では，トランザクションを投機的に並列実行することで，ロックに比べ並列度が向上する．しかし，HTM では同一共有変数へのアクセス競合が頻発することによる性能低下が問題となる．この問題に対し，トランザクションのスケジューリングを改良することにより競合の発生を抑制する研究が多く行われてきたが，そのいずれの手法を用いても，十分な性能向上が得られていないプログラムが存在する．そこで本稿ではまず，そのようなプログラムが持つメモリアクセスパターンを調査し，性能向上が妨げられている原因を調査した．その結果，複数の実行フェーズを持ち，あるフェーズでアクセスした共有変数に対し，以降のフェーズで再度アクセスしないようなトランザクションが存在することが分かった．そこで本稿では，トランザクションが持つ実行フェーズを考慮して競合検出を行うことで性能を向上させる手法を提案する．Contention および Deque を用いて評価を行った結果，Contention で平均 63.2%，Deque で平均 6.3% の性能向上を確認した．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"9","bibliographic_titles":[{"bibliographic_title":"研究報告システム・アーキテクチャ（ARC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2017-07-19","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"40","bibliographicVolumeNumber":"2017-ARC-227"}]},"relation_version_is_last":true,"weko_creator_id":"11"}}