{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00235625","sets":["6504:11678:11698"]},"path":["11698"],"owner":"44499","recid":"235625","title":["OpenACCを用いた単精度LU分解のGPU並列化"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-03-01"},"_buckets":{"deposit":"1997482b-bf1a-4522-862f-8c5c42fac527"},"_deposit":{"id":"235625","pid":{"type":"depid","value":"235625","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"OpenACCを用いた単精度LU分解のGPU並列化","author_link":["643896","643898","643895","643897","643899"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"OpenACCを用いた単精度LU分解のGPU並列化"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"コンピュータシステム","subitem_subject_scheme":"Other"}]},"item_type_id":"22","publish_date":"2024-03-01","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_22_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"工学院大"},{"subitem_text_value":"工学院大"},{"subitem_text_value":"工学院大"},{"subitem_text_value":"工学院大"},{"subitem_text_value":"工学院大"}]},"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/235625/files/IPSJ-Z86-1J-05.pdf","label":"IPSJ-Z86-1J-05.pdf"},"date":[{"dateType":"Available","dateValue":"2024-07-03"}],"format":"application/pdf","filename":"IPSJ-Z86-1J-05.pdf","filesize":[{"value":"305.4 kB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"81163a83-6e9e-448f-a4a8-a31ba88c2b45","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_22_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"高山, 智礼"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"久木田, 仁"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"矢島, 雄河"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"藤井, 昭宏"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田中, 輝雄"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_22_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00349328","subitem_source_identifier_type":"NCID"}]},"item_22_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"機械学習の分野では大容量の学習データ自体の精度が低いため，GPUの単精度（あるいは半精度）演算の高い計算能力が注目されている．しかし，GPUの取り扱いにCUDAを用いるとCPU向けのプログラムを大幅に書き換える必要がある．そのため書き換えを大幅に削減可能な，ディレクティブベースのGPU向けプログラミングモデルOpenACCによる利用が広がっている．一方，スーパーコンピュータのベンチマークHPL-MxPでは混合精度演算が用いられており，ここでの主な演算である密行列のLU分解も単精度（あるいは半精度）演算である．本研究では， OpenACCを用いて，HPL-MxP内のLU分解(単精度)をGPU上で並列化し，CPU上での実行を越える性能を得た．","subitem_description_type":"Other"}]},"item_22_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"28","bibliographic_titles":[{"bibliographic_title":"第86回全国大会講演論文集"}],"bibliographicPageStart":"27","bibliographicIssueDates":{"bibliographicIssueDate":"2024-03-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2024"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":235625,"updated":"2025-01-19T09:35:09.017155+00:00","links":{},"created":"2025-01-19T01:37:12.334647+00:00"}