{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00206324","sets":["1164:2240:10178:10282"]},"path":["10282"],"owner":"44499","recid":"206324","title":["SX-Aurora TSUBASAにおける有限要素解析のための共役勾配法の性能評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2020-07-23"},"_buckets":{"deposit":"53101b71-cb85-4f52-95a5-c768c6d1e2d9"},"_deposit":{"id":"206324","pid":{"type":"depid","value":"206324","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"SX-Aurora TSUBASAにおける有限要素解析のための共役勾配法の性能評価","author_link":["513050","513054","513053","513049","513051","513052","513048"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"SX-Aurora TSUBASAにおける有限要素解析のための共役勾配法の性能評価"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"高性能計算","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2020-07-23","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"株式会社科学計算総合研究所"},{"subitem_text_value":"株式会社科学計算総合研究所"},{"subitem_text_value":"一社）インダストリスパコン推進センター"},{"subitem_text_value":"一社）インダストリスパコン推進センター"},{"subitem_text_value":"日本電気(株)AIプラットフォーム事業部"},{"subitem_text_value":"日本電気(株)AIプラットフォーム事業部"},{"subitem_text_value":"東京大学大学院新領域創成科学研究科"}]},"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/206324/files/IPSJ-HPC20175018.pdf","label":"IPSJ-HPC20175018.pdf"},"date":[{"dateType":"Available","dateValue":"2022-07-23"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-HPC20175018.pdf","filesize":[{"value":"3.0 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":"acd91dac-31f5-4bd1-b34d-0b8e182c4035","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2020 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"菱沼, 利彰"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"井原, 遊"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"高村, 守幸"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"平野, 哲"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"萩原, 孝"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"岩田, 直樹"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"奥田, 洋司"}],"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":"有限要素法を用いた構造解析は工学分野で広く利用されている．一般的に有限要素法を用いたシミュレーションにおいて最も時間がかかるのは疎行列を係数行列としてもつ連立一次方程式の求解である．近年，NEC よりベクトル型のアクセラレータボード (Vector Engine) を搭載した SX-Aurora TSUBASA が登場した．Vector Engine には 1 ボードあたり約 1.2 TB/s の帯域をもつ高速メモリと 8 つの高性能演算コアが搭載されており，各コアには倍精度 32 要素を同時演算できる FMA 演算器が 3 器搭載されている．有限要素解析は PC クラスタをはじめとするスカラプロセッサ・システム向けの高速化研究が進んでいるが，ベクトル型アクセラレータボード上における性能や有効性の評価は行われていない．本研究では有限要素解析ソフトウェアである FrontISTR に含まれる線形ソルバから疎行列の格納形式として BCRS 形式と JAD 形式，線形ソルバとして CG 法を選び，格納形式や実装ごとの OpenMP によるマルチスレッド化，およびコンパイラの自動ベクトル化による高速化効果や性能について評価した．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"10","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":"2020-07-23","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"18","bibliographicVolumeNumber":"2020-HPC-175"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":206324,"updated":"2025-01-19T19:27:25.489190+00:00","links":{},"created":"2025-01-19T01:08:23.686511+00:00"}