{"updated":"2025-01-21T12:25:44.004121+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00098527","sets":["581:7397:7450"]},"path":["7450"],"owner":"11","recid":"98527","title":["GPUオンチップマイクロコントローラへのデータ転送処理オフロード"],"pubdate":{"attribute_name":"公開日","attribute_value":"2014-02-15"},"_buckets":{"deposit":"ca55fd64-23f0-4e0e-a206-a69a61243402"},"_deposit":{"id":"98527","pid":{"type":"depid","value":"98527","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"GPUオンチップマイクロコントローラへのデータ転送処理オフロード","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"GPUオンチップマイクロコントローラへのデータ転送処理オフロード"},{"subitem_title":"Data Transfer Processing Offload to the GPU On-chip Microcontrollers","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[特集：組込みシステム工学] GPU，GPGPU，GPUデータ転送，マイクロコントローラ，開発環境","subitem_subject_scheme":"Other"}]},"item_type_id":"2","publish_date":"2014-02-15","item_2_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"立命館大学大学院情報理工学研究科"},{"subitem_text_value":"立命館大学情報理工学部"},{"subitem_text_value":"立命館大学情報理工学部"},{"subitem_text_value":"名古屋大学大学院情報科学研究科"}]},"item_2_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science and Engineering, Ritsumeikan University","subitem_text_language":"en"},{"subitem_text_value":"College of Information Science and Engineering, Ritsumeikan Uniersity","subitem_text_language":"en"},{"subitem_text_value":"College of Information Science and Engineering, Ritsumeikan Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science, Nagoya Uniersity","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"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/98527/files/IPSJ-JNL5502051.pdf"},"date":[{"dateType":"Available","dateValue":"2016-02-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-JNL5502051.pdf","filesize":[{"value":"2.4 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":"8"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"66bd1fcf-b76a-4816-bb43-bbb21036ac8a","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2014 by the Information Processing Society of Japan"}]},"item_2_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"藤居, 祐輔"},{"creatorName":"安積, 卓也"},{"creatorName":"西尾, 信彦"},{"creatorName":"加藤, 真平"}],"nameIdentifiers":[{}]}]},"item_2_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yusuke, Fujii","creatorNameLang":"en"},{"creatorName":"Takuya, Azumi","creatorNameLang":"en"},{"creatorName":"Nobuhiko, Nishio","creatorNameLang":"en"},{"creatorName":"Shinpei, Kato","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_2_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00116647","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_2_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7764","subitem_source_identifier_type":"ISSN"}]},"item_2_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"サイバーフィジカルシステム（CPS）が注目される中，その技術基盤として，GPUなどのデバイスが利用され始めている．GPUはデバイスドライバを経由して利用されているが，CPSのように短い周期で繰り返し多くの処理が行われると，ホストへの負担が増えるとともに，デバイス制御や処理の同期によってレイテンシが発生する．さらにGPU処理では，データをデバイスメモリへと転送する必要があり，上記問題を悪化させ，データ転送処理自体にも影響を与える．そのため我々は，GPU制御処理の一部をGPUマイクロコントローラ上で動作するファームウェアへオフロードし，GPU処理の効率化をめざす．本論文では，オフロード基盤としてコンパイラ，デバッグ支援ツールを含んだGPU制御ファームウェア開発環境と，既存のNVIDIA社製ファームウェアと同等の機能を持つファームウェアを開発する．次に，オフロード基盤を用いて，制御処理の一部であるDMA転送処理をファームウェアに追加実装することで，オフロードを実現しGPU処理を効率化する．我々は，実装したファームウェアと既存のファームウェアを比較し，性能低下がないことを示すことで，オフロード基盤の有効性を確認した．オフロードしたデータ転送処理では，既存のデータ転送処理と比べ，一部のデータサイズにおいて約1.5倍の転送速度を実現し，さらに既存データ転送処理へのオーバラップ転送を実現した．","subitem_description_type":"Other"}]},"item_2_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Cyber-Physical Systems (CPS) aim to control complex real-world phenomenon. The computational cost and real-time constraints could be a major challenge of CPS. GPUs have come to be used as base technology for CPS. CPS applications have many short periodic tasks. Therefore latency is occurred and CPU load is increased because of synchronizations and control processes when the CPS application runs on a GPU. In this paper, we present an offloading data transfer process to GPU microcontrollers. We aim at effective GPU processing which reduce the CPU load and improve the data transfer. First of all, we prepared the process-offloading infrastructure, which has the GPU microcontrollers firmware development environment and the no-extensible firmware. Secondly, we provide to data transfer methods by implementing an extensible firmware; transferring data by microcontrollers and transferring data by overlapping to the other data transfers and microcontrollers. Our extensible firmwares do not have performance degradation from the performance of existing firmwares. In addition, we compared our data transfer methods with existing methods. As a result, our transfer methods are one and a half times as fast as transfer speed of existing method when the data size band of part.","subitem_description_type":"Other"}]},"item_2_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"1058","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌"}],"bibliographicPageStart":"1048","bibliographicIssueDates":{"bibliographicIssueDate":"2014-02-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"55"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-18T23:44:45.445567+00:00","id":98527,"links":{}}