{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00217101","sets":["1164:1579:10818:10892"]},"path":["10892"],"owner":"44499","recid":"217101","title":["Layer-wise power/performance modelling for single-board CNN inference"],"pubdate":{"attribute_name":"公開日","attribute_value":"2022-03-03"},"_buckets":{"deposit":"cf8b0ff0-e093-4039-b097-3ed6e0cb8df4"},"_deposit":{"id":"217101","pid":{"type":"depid","value":"217101","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"Layer-wise power/performance modelling for single-board CNN inference","author_link":["561770","561769","561776","561773","561767","561774","561775","561771","561772","561768"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Layer-wise power/performance modelling for single-board CNN inference"},{"subitem_title":"Layer-wise power/performance modelling for single-board CNN inference","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ニューラルネットワーク","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2022-03-03","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"Kyushu Uniersity"},{"subitem_text_value":"Kyushu Uniersity"},{"subitem_text_value":"Kyushu Uniersity"},{"subitem_text_value":"Kyushu Uniersity"},{"subitem_text_value":"Kyushu Uniersity"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Kyushu Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Kyushu Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Kyushu Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Kyushu Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Kyushu Uniersity","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"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/217101/files/IPSJ-ARC22248013.pdf","label":"IPSJ-ARC22248013.pdf"},"date":[{"dateType":"Available","dateValue":"2024-03-03"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-ARC22248013.pdf","filesize":[{"value":"20.9 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":"16"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"7391fea1-fded-4d87-85ed-8c5b736e6b4e","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2022 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yi, Ng Kuan"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Aalaa, M.A. Babai"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Satoshi, Kawakami"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Teruo, Tanimoto"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Koji, Inoue"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yi, Ng Kuan","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Aalaa, M.A. Babai","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Satoshi, Kawakami","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Teruo, Tanimoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Koji, Inoue","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":"Intermittent executions and energy harvesting technologies are promising candidates to enable renewable energy on small-scale computer systems like single-board computers, making sustainable computing possible. In this work, we implemented an energy consumption prediction framework for each layer of CNN executing on single-board computers based on NeuralPower as the first step towards enabling energy-efficient intermittent execution of CNN inference on single-board computers. We found that layer hyperparameters cannot explain all the variations in execution time and power consumption when the layer is executed. Model's prediction can be improved with the knowledge of performance counter values, but these values are not available before a layer is executed. Furthermore, our analysis revealed that implementation optimization like sparse matrix multiplication might cause a layer's execution time and power to change with its input values.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Intermittent executions and energy harvesting technologies are promising candidates to enable renewable energy on small-scale computer systems like single-board computers, making sustainable computing possible. In this work, we implemented an energy consumption prediction framework for each layer of CNN executing on single-board computers based on NeuralPower as the first step towards enabling energy-efficient intermittent execution of CNN inference on single-board computers. We found that layer hyperparameters cannot explain all the variations in execution time and power consumption when the layer is executed. Model's prediction can be improved with the knowledge of performance counter values, but these values are not available before a layer is executed. Furthermore, our analysis revealed that implementation optimization like sparse matrix multiplication might cause a layer's execution time and power to change with its input values.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"11","bibliographic_titles":[{"bibliographic_title":"研究報告システム・アーキテクチャ（ARC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2022-03-03","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"13","bibliographicVolumeNumber":"2022-ARC-248"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":217101,"updated":"2025-01-19T15:37:19.400062+00:00","links":{},"created":"2025-01-19T01:17:36.444708+00:00"}