{"created":"2025-01-19T01:10:47.312407+00:00","updated":"2025-01-19T18:30:30.971607+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00209481","sets":["934:1119:10468:10469"]},"path":["10469"],"owner":"44499","recid":"209481","title":["Power Prediction for Sustainable HPC"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-02-17"},"_buckets":{"deposit":"77d3b08d-7b7c-441a-8749-ce26b0d6d192"},"_deposit":{"id":"209481","pid":{"type":"depid","value":"209481","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"Power Prediction for Sustainable HPC","author_link":["528020","528021","528017","528016","528022","528023","528024","528026","528014","528027","528028","528029","528025","528030","528018","528031","528015","528019"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Power Prediction for Sustainable HPC"},{"subitem_title":"Power Prediction for Sustainable HPC","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"power prediction, HPC, deep learning, job scheduling","subitem_subject_scheme":"Other"}]},"item_type_id":"3","publish_date":"2021-02-17","item_3_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"Fujitsu Laboratories LTD."},{"subitem_text_value":"Fujitsu LTD."},{"subitem_text_value":"Fujitsu Laboratories LTD."},{"subitem_text_value":"Fujitsu Laboratories LTD."},{"subitem_text_value":"Fujitsu Laboratories LTD."},{"subitem_text_value":"Fujitsu Laboratories LTD."},{"subitem_text_value":"Fujitsu LTD."},{"subitem_text_value":"Fujitsu LTD."},{"subitem_text_value":"RIKEN Center for Computational Science"}]},"item_3_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Fujitsu Laboratories LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu Laboratories LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu Laboratories LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu Laboratories LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu Laboratories LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu LTD.","subitem_text_language":"en"},{"subitem_text_value":"Fujitsu LTD.","subitem_text_language":"en"},{"subitem_text_value":"RIKEN Center for Computational Science","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/209481/files/IPSJ-TACS1401003.pdf","label":"IPSJ-TACS1401003.pdf"},"date":[{"dateType":"Available","dateValue":"2023-02-17"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-TACS1401003.pdf","filesize":[{"value":"2.4 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"5"},{"tax":["include_tax"],"price":"0","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"16"},{"tax":["include_tax"],"price":"0","billingrole":"11"},{"tax":["include_tax"],"price":"0","billingrole":"14"},{"tax":["include_tax"],"price":"0","billingrole":"15"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"7c5d8e62-295f-4528-928f-6eea0c98eaf1","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Information Processing Society of Japan"}]},"item_3_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Shigeto, Suzuki"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Michiko, Hiraoka"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Takashi, Shiraishi"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Enxhi, Kreshpa"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Takuji, Yamamoto"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroyuki, Fukuda"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shuji, Matsui"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masahide, Fujisaki"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Atsuya, Uno"}],"nameIdentifiers":[{}]}]},"item_3_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Shigeto, Suzuki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Michiko, Hiraoka","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Takashi, Shiraishi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Enxhi, Kreshpa","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Takuji, Yamamoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroyuki, Fukuda","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shuji, Matsui","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masahide, Fujisaki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Atsuya, Uno","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_3_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11833852","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_3_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7829","subitem_source_identifier_type":"ISSN"}]},"item_3_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Exascale computers consume huge amounts of power and their variation over time makes system energy management important. Because of time lag in cooling-units operation, predictive control is desirable for effective power control. In this work, we report a state-of-the-art power prediction model. Conventional methods with topic model use the power of past job as a prediction based on the similarity of job information. The prediction, however, fails, if there is no correct data before. To resolve this, we developed a recurrent neural network model with variable network size, which detects features of power shape from its power history and enables precise prediction during job execution. By integrating these models into a single algorithm, the optimal model is automatically adopted for prediction according to the job status. We demonstrated high-precision prediction with an average relative error of 5.7% in K computer as compared to that of 20.1% by the conventional method.\n------------------------------\nThis is a preprint of an article intended for publication Journal of\nInformation Processing(JIP). This preprint should not be cited. This\narticle should be cited as: Journal of Information Processing Vol.29(2021) (online)\n------------------------------","subitem_description_type":"Other"}]},"item_3_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Exascale computers consume huge amounts of power and their variation over time makes system energy management important. Because of time lag in cooling-units operation, predictive control is desirable for effective power control. In this work, we report a state-of-the-art power prediction model. Conventional methods with topic model use the power of past job as a prediction based on the similarity of job information. The prediction, however, fails, if there is no correct data before. To resolve this, we developed a recurrent neural network model with variable network size, which detects features of power shape from its power history and enables precise prediction during job execution. By integrating these models into a single algorithm, the optimal model is automatically adopted for prediction according to the job status. We demonstrated high-precision prediction with an average relative error of 5.7% in K computer as compared to that of 20.1% by the conventional method.\n------------------------------\nThis is a preprint of an article intended for publication Journal of\nInformation Processing(JIP). This preprint should not be cited. This\narticle should be cited as: Journal of Information Processing Vol.29(2021) (online)\n------------------------------","subitem_description_type":"Other"}]},"item_3_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌コンピューティングシステム（ACS）"}],"bibliographicIssueDates":{"bibliographicIssueDate":"2021-02-17","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"14"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":209481,"links":{}}