{"created":"2025-01-19T01:35:45.156682+00:00","updated":"2025-01-19T09:54:05.248463+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00234031","sets":["1164:2240:11467:11597"]},"path":["11597"],"owner":"44499","recid":"234031","title":["GPUにおける8倍精度高速フーリエ変換の実装と評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-05-01"},"_buckets":{"deposit":"f80fc9de-fc39-44a1-83a7-d967ab23b3c0"},"_deposit":{"id":"234031","pid":{"type":"depid","value":"234031","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"GPUにおける8倍精度高速フーリエ変換の実装と評価","author_link":["636991","636992"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"GPUにおける8倍精度高速フーリエ変換の実装と評価"},{"subitem_title":"Implementation and Evaluation of Octuple Precision Fast Fourier Transform on a GPU","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"数値計算","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-05-01","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"筑波大学大学院理工情報生命学術院"},{"subitem_text_value":"筑波大学計算科学研究センター"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Science and Technology, University of Tsukuba","subitem_text_language":"en"},{"subitem_text_value":"Center for Computational Sciences, University of Tsukuba","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/234031/files/IPSJ-HPC24194002.pdf","label":"IPSJ-HPC24194002.pdf"},"date":[{"dateType":"Available","dateValue":"2026-05-01"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-HPC24194002.pdf","filesize":[{"value":"1.6 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":"73eaca67-261c-415f-9592-177174804728","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"川上, 昌汰"}],"nameIdentifiers":[{}]},{"creatorNames":[{"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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8841","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"高速フーリエ変換（Fast Fourier Transform，FFT）は周波数解析や常微分方程式の数値計算，多倍長乗算などに用いられる重要なアルゴリズムである．一方，大規模な計算を行う際，一般的な計算機で実装されている倍精度演算では精度が不足する場合がある．この問題を解決するために，倍精度演算以上の精度で計算することができる高精度演算が用いられる．本研究では，倍精度浮動小数点数（double 型）を 4 個使用して，8 倍精度浮動小数点数を表現するQD（Quad-Double）型演算を用いて 8 倍精度 FFT を実装した．加えて，OpenMP を用いて GPU（Graphic Processing Unit）で並列化を行った．同様に OpenMP を用いて CPU で並列化を行った実装と比較し，最大約 7.84 倍の高速化を実現した．また，倍精度 FFT と比べ，QD 型演算を用いた 8 倍精度 FFT は 166.2 倍の演算量を必要とするにもかかわらず，GPU で並列化を行った 8 倍精度 FFT は，cuFFT による倍精度 FFT と比べ，約 8.85 倍から約 100 倍の実行時間で計算可能であった．","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":"2024-05-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"2024-HPC-194"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":234031,"links":{}}