{"id":"https://openalex.org/W7124145190","doi":"https://doi.org/10.1109/icpads67057.2025.11323149","title":"AUE: A Normalized Energy Efficiency Metric for AI Servers Under LLM Workloads","display_name":"AUE: A Normalized Energy Efficiency Metric for AI Servers Under LLM Workloads","publication_year":2025,"publication_date":"2025-12-14","ids":{"openalex":"https://openalex.org/W7124145190","doi":"https://doi.org/10.1109/icpads67057.2025.11323149"},"language":null,"primary_location":{"id":"doi:10.1109/icpads67057.2025.11323149","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpads67057.2025.11323149","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101712600","display_name":"Yijia Zhang","orcid":"https://orcid.org/0000-0001-6925-7777"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yijia Zhang","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032211015","display_name":"Dongxiang Zhang","orcid":"https://orcid.org/0009-0006-6338-0698"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongxiang Zhang","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101713527","display_name":"Xianglin Liu","orcid":"https://orcid.org/0000-0001-8883-7056"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xianglin Liu","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123007085","display_name":"Qiang Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiang Wang","raw_affiliation_strings":["Harbin Institute of Technology,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,Shenzhen,China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102768672","display_name":"Bingqiang Wang","orcid":"https://orcid.org/0009-0004-4964-3258"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bingqiang Wang","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036667746","display_name":"Shixun Zhang","orcid":"https://orcid.org/0000-0001-9487-4697"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shixun Zhang","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023918894","display_name":"Yonghong Tian","orcid":"https://orcid.org/0000-0002-2978-5935"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yonghong Tian","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12238","display_name":"Green IT and Sustainability","score":0.3513999879360199,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12238","display_name":"Green IT and Sustainability","score":0.3513999879360199,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.323199987411499,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.09740000218153,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.7360000014305115},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.720300018787384},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.5784000158309937},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.5658000111579895},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5604000091552734},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5081999897956848},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4106000065803528}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7573000192642212},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.7360000014305115},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.720300018787384},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.5784000158309937},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.5658000111579895},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5604000091552734},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5081999897956848},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4106000065803528},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.3959999978542328},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.3666999936103821},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3598000109195709},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.31439998745918274},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31189998984336853},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C3018263672","wikidata":"https://www.wikidata.org/wiki/Q1296251","display_name":"Efficient algorithm","level":2,"score":0.27379998564720154},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2728999853134155},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2669000029563904},{"id":"https://openalex.org/C2986957236","wikidata":"https://www.wikidata.org/wiki/Q8036562","display_name":"Energy resources","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C2778648169","wikidata":"https://www.wikidata.org/wiki/Q967768","display_name":"Compatibility (geochemistry)","level":2,"score":0.25940001010894775},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.2547999918460846}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icpads67057.2025.11323149","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpads67057.2025.11323149","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8305507302284241,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2169631286","https://openalex.org/W3157657667","https://openalex.org/W3204998121","https://openalex.org/W4387494819","https://openalex.org/W4390189088","https://openalex.org/W4394923310","https://openalex.org/W4394945061","https://openalex.org/W4399074329","https://openalex.org/W4399449628","https://openalex.org/W4402896949","https://openalex.org/W4407197043","https://openalex.org/W4409248734","https://openalex.org/W4409248825","https://openalex.org/W4411345920","https://openalex.org/W4411346147"],"related_works":[],"abstract_inverted_index":{"Under":[0],"the":[1,9,33,60,103,121,131,194,211,220],"rapid":[2],"advancement":[3],"of":[4,13,37,62,124,133,202],"large":[5],"model-driven":[6],"artificial":[7],"intelligence,":[8],"surging":[10],"energy":[11,28,97,104,122,178,252],"consumption":[12],"AI":[14,63,96,251],"training":[15,159],"and":[16,26,35,47,70,85,135,154,160,233,237,249],"inference":[17,161],"tasks":[18],"has":[19],"created":[20],"an":[21],"urgent":[22],"need":[23],"for":[24,74,210],"precise":[25],"comparable":[27],"efficiency":[29,98,123,253],"metrics":[30,43,49,67,169],"to":[31,79,191],"guide":[32],"design":[34],"deployment":[36],"green":[38],"computing":[39],"systems.":[40],"While":[41],"existing":[42],"such":[44],"as":[45,102],"PUE":[46],"Green500":[48],"focus":[50],"on":[51,138],"infrastructure":[52],"or":[53],"traditional":[54,167],"numerical":[55],"computations,":[56],"they":[57,76],"cannot":[58],"reflect":[59],"characteristics":[61],"workloads.":[64,162],"Although":[65],"applicationoriented":[66],"like":[68],"J/response":[69],"J/token":[71,168,188],"are":[72],"designed":[73],"LLMs,":[75],"remain":[77],"susceptible":[78],"biases":[80],"induced":[81],"by":[82],"model":[83,115,196,222,238],"scale":[84],"output":[86],"strategies,":[87],"lacking":[88],"cross-model":[89],"comparability.":[90],"This":[91],"paper":[92],"proposes":[93],"a":[94,139,198,242,247],"novel":[95],"metric,":[99],"AUE,":[100],"defined":[101],"consumed":[105],"per":[106,109],"thousand":[107],"tokens":[108],"billion":[110],"activated":[111],"parameters.":[112],"By":[113],"normalizing":[114],"size":[116],"effects,":[117],"AUE":[118,134,175,201,240],"accurately":[119],"reflects":[120],"underlying":[125],"computational":[126],"resources.":[127],"We":[128],"theoretically":[129],"justify":[130],"validity":[132],"conduct":[136],"experiments":[137],"server":[140],"equipped":[141],"with":[142,216],"<tex":[143],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[144],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$4":[145],"\\times$</tex>":[146],"Ascend":[147],"NPU":[148],"910C":[149],"accelerators,":[150],"evaluating":[151],"dense":[152],"Transformer":[153],"MoE":[155],"architectures":[156],"across":[157,235],"both":[158],"Experimental":[163],"results":[164],"demonstrate":[165],"that":[166],"disproportionately":[170],"favor":[171],"smaller":[172,225],"models,":[173],"whereas":[174],"reveals":[176],"true":[177],"utilization":[179],"efficiency.":[180],"For":[181],"instance,":[182],"while":[183],"Qwen3":[184,192],"0.6B":[185],"shows":[186],"superior":[187],"values":[189],"compared":[190],"14B,":[193],"14B":[195,221],"achieves":[197],"significantly":[199],"better":[200],"12.88":[203],"J/(KToken":[204,208],"GParam)":[205,209],"versus":[206],"24.35":[207],"0.6":[212],"B":[213],"model,":[214],"consistent":[215],"measured":[217],"FLOPs":[218],"where":[219],"outperforms":[223],"its":[224],"counterpart.":[226],"With":[227],"advantages":[228],"including":[229],"simple":[230],"measurement":[231],"procedures":[232],"compatibility":[234],"platforms":[236],"architectures,":[239],"provides":[241],"viable":[243],"pathway":[244],"toward":[245],"establishing":[246],"normalized":[248],"standardized":[250],"evaluation":[254],"framework.":[255]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-01-15T00:00:00"}