{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00231066","sets":["1164:1579:11081:11407"]},"path":["11407"],"owner":"44499","recid":"231066","title":["MN-Coreアクセラレータ上でのBLASの実装と評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-11-28"},"_buckets":{"deposit":"a7c2cdce-f0c3-44f8-b93a-e019c663223a"},"_deposit":{"id":"231066","pid":{"type":"depid","value":"231066","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"MN-Coreアクセラレータ上でのBLASの実装と評価","author_link":["623224","623223","623221","623222"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"MN-Coreアクセラレータ上でのBLASの実装と評価"},{"subitem_title":"Implementation and evaluation of BLAS on MN-Core accelerator","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"アーキテクチャ","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2023-11-28","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"株式会社Preferred Networks"},{"subitem_text_value":"東京大学情報理工学系研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Preferred Networks, Inc.","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","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/231066/files/IPSJ-ARC23255027.pdf","label":"IPSJ-ARC23255027.pdf"},"date":[{"dateType":"Available","dateValue":"2025-11-28"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-ARC23255027.pdf","filesize":[{"value":"329.8 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":"4319e2e7-c1fc-4a52-b6de-f59b897e03e9","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"井町, 宏人"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"野崎, 愛"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hiroto, Imachi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ai, Nozaki","creatorNameLang":"en"}],"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":"ディープラーニング向けアクセラレータである MN-Core 上で倍精度の BLAS ルーチンを一部を除き実装し性能評価を行った．アクセラレータ内のデータ移動を最小化するデータレイアウトの考慮やレジスタ割り当ての最適化により，大部分のルーチンで実効効率が原理的な上限に漸近することを確認した．実装と性能評価の詳細を報告する．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"We implemented a part of double precision BLAS routines on MN-Core, an accelerator for deep learning, and evaluate their performance. Most of the implemented routines show efficiency approaching their upper limit after optimizing data layouts and register allocation to minimize data movement. Details of implementation and performance evaluation are reported.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告システム・アーキテクチャ（ARC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2023-11-28","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"27","bibliographicVolumeNumber":"2023-ARC-255"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:31:12.433172+00:00","updated":"2025-01-19T10:53:21.749595+00:00","id":231066}