{"id":90198,"created":"2025-01-18T23:39:49.414247+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00090198","sets":["1164:2240:7073:7075"]},"path":["7075"],"owner":"11","recid":"90198","title":["GPUにおける高速なCRS形式疎行列ベクトル積の実装"],"pubdate":{"attribute_name":"公開日","attribute_value":"2013-02-14"},"_buckets":{"deposit":"1350ab0d-c82e-434a-b6d9-145317e5fc59"},"_deposit":{"id":"90198","pid":{"type":"depid","value":"90198","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"GPUにおける高速なCRS形式疎行列ベクトル積の実装","author_link":["0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"GPUにおける高速なCRS形式疎行列ベクトル積の実装"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"数値ライブラリ","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2013-02-14","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"筑波大学大学院システム情報工学研究科"},{"subitem_text_value":"筑波大学システム情報系"}]},"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/90198/files/IPSJ-HPC13138005.pdf"},"date":[{"dateType":"Available","dateValue":"2015-02-14"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-HPC13138005.pdf","filesize":[{"value":"1.4 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":"14"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"6271e3ce-9c84-448b-8279-4bd0ecba4b81","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2013 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"椋木大地"},{"creatorName":"高橋大介"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10463942","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_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"疎行列ベクトル積 （SpMV） は科学技術計算において多用される重要な基本演算である．本稿では GPU における高速な CRS 形式 SpMV の実装について報告する．GPU として NVIDIA 社の Kepler アーキテクチャを対象とし，CUDA5.0 環境において実装を行った．従来の Fermi アーキテクチャまでの GPU を対象に提案されていた実装手法をベースに，Kepler アーキテクチャで新たにサポートされた機能や仕様変更を活用して，最適化を行った．Kepler アーキテクチャの Tesla K20 における性能評価では，CUDA5.0 に付属の cuSPARSE における CRS 形式の倍精度 SpMV ルーチンに対して，200 種類の行列において，平均で約 1.86 倍，177 種類の行列で性能向上を達成した．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"7","bibliographic_titles":[{"bibliographic_title":"研究報告ハイパフォーマンスコンピューティング（HPC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2013-02-14","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"5","bibliographicVolumeNumber":"2013-HPC-138"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"updated":"2025-01-21T16:08:14.796527+00:00","links":{}}