{"id":"https://openalex.org/W4394945233","doi":"https://doi.org/10.1145/3627703.3629572","title":"CDMPP: A Device-Model Agnostic Framework for Latency Prediction of Tensor Programs","display_name":"CDMPP: A Device-Model Agnostic Framework for Latency Prediction of Tensor Programs","publication_year":2024,"publication_date":"2024-04-18","ids":{"openalex":"https://openalex.org/W4394945233","doi":"https://doi.org/10.1145/3627703.3629572"},"language":"en","primary_location":{"id":"doi:10.1145/3627703.3629572","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627703.3629572","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Nineteenth European Conference on Computer Systems","raw_type":"proceedings-article"},"type":"article","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/A5095776670","display_name":"Hanpeng Hu","orcid":"https://orcid.org/0009-0008-5787-5226"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Hanpeng Hu","raw_affiliation_strings":["University of Hong Kong, Hong Kong and ByteDance Inc"],"raw_orcid":"https://orcid.org/0009-0008-5787-5226","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong and ByteDance Inc","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101529879","display_name":"Junwei Su","orcid":"https://orcid.org/0000-0003-0537-4004"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Junwei Su","raw_affiliation_strings":["University of Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0003-0537-4004","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087869611","display_name":"Juntao Zhao","orcid":"https://orcid.org/0000-0003-3376-0607"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Juntao Zhao","raw_affiliation_strings":["University of Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0003-3376-0607","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013856812","display_name":"Yanghua Peng","orcid":"https://orcid.org/0000-0003-3989-4358"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanghua Peng","raw_affiliation_strings":["ByteDance Inc., Seattle, Washington, USA"],"raw_orcid":"https://orcid.org/0000-0003-3989-4358","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, Washington, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028120333","display_name":"Yibo Zhu","orcid":"https://orcid.org/0000-0002-9113-2660"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yibo Zhu","raw_affiliation_strings":["ByteDance Inc., Seattle, Washington, USA"],"raw_orcid":"https://orcid.org/0000-0002-9113-2660","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, Washington, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084047386","display_name":"Haibin Lin","orcid":"https://orcid.org/0000-0003-4879-5335"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haibin Lin","raw_affiliation_strings":["ByteDance Inc., Mountain View, California, USA"],"raw_orcid":"https://orcid.org/0000-0003-4879-5335","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Mountain View, California, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012597518","display_name":"Chuan Wu","orcid":"https://orcid.org/0000-0002-3144-4398"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chuan Wu","raw_affiliation_strings":["University of Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-3144-4398","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5095776670"],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":1.8974,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.85525825,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1054","last_page":"1074"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9950000047683716,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9950000047683716,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9945999979972839,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T12303","display_name":"Tensor decomposition and applications","score":0.9797999858856201,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8051491975784302},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.725395679473877},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.6303660869598389},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5586853623390198},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5165168642997742},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5089480876922607},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.45132485032081604},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.41213545203208923},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3844478130340576},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.3510301113128662}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8051491975784302},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.725395679473877},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.6303660869598389},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5586853623390198},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5165168642997742},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5089480876922607},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.45132485032081604},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.41213545203208923},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3844478130340576},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3510301113128662},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3627703.3629572","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627703.3629572","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Nineteenth European Conference on Computer Systems","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W129305155","https://openalex.org/W1686810756","https://openalex.org/W1728842521","https://openalex.org/W2763068163","https://openalex.org/W2770501762","https://openalex.org/W2804032941","https://openalex.org/W2884700152","https://openalex.org/W2942889782","https://openalex.org/W2961619211","https://openalex.org/W2962746461","https://openalex.org/W2962897394","https://openalex.org/W2964248614","https://openalex.org/W2964278684","https://openalex.org/W2964330541","https://openalex.org/W2970971581","https://openalex.org/W2978110818","https://openalex.org/W2990200213","https://openalex.org/W3035582633","https://openalex.org/W3037749908","https://openalex.org/W3043571714","https://openalex.org/W3047049572","https://openalex.org/W3093557617","https://openalex.org/W3119322020","https://openalex.org/W3139325691","https://openalex.org/W3157020554","https://openalex.org/W3163725856","https://openalex.org/W3165698711","https://openalex.org/W3181228931","https://openalex.org/W3188978989","https://openalex.org/W3207748736","https://openalex.org/W3214511341","https://openalex.org/W4226220908","https://openalex.org/W4287634497","https://openalex.org/W4318541553","https://openalex.org/W4320067919","https://openalex.org/W6687483927","https://openalex.org/W6713134421","https://openalex.org/W6713955831","https://openalex.org/W6739901393","https://openalex.org/W6752057402","https://openalex.org/W6756040250","https://openalex.org/W6769062451","https://openalex.org/W6780827055","https://openalex.org/W6786080308","https://openalex.org/W6869608176","https://openalex.org/W6987026237"],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W4321636575","https://openalex.org/W1986418932","https://openalex.org/W2357796999","https://openalex.org/W2045526782","https://openalex.org/W2741131631","https://openalex.org/W2156919374","https://openalex.org/W2307385607","https://openalex.org/W3128807919","https://openalex.org/W3176411177"],"abstract_inverted_index":{"Deep":[0],"Neural":[1],"Networks":[2],"(DNNs)":[3],"have":[4,85],"shown":[5],"excellent":[6],"performance":[7,72,95],"in":[8,33],"a":[9,21,28,67,87,137,152,161,183],"wide":[10],"range":[11,185],"of":[12,19,50,59,73,81,96,130,147,186,213],"machine":[13],"learning":[14],"applications.":[15],"Knowing":[16],"the":[17,47,71,82,94,144,166,221],"latency":[18,114],"running":[20],"DNN":[22,38,51,74,175,187],"model":[23,70,89],"or":[24,40],"tensor":[25,98,112,131,148],"program":[26,113],"on":[27,65,76,182],"specific":[29],"device":[30,44],"is":[31],"useful":[32],"various":[34,97],"tasks,":[35],"such":[36],"as":[37],"graph-":[39],"tensor-level":[41],"optimization":[42],"and":[43,53,104,120,136,159,177,189,200,206,210,220],"selection.":[45],"Considering":[46],"large":[48],"space":[49],"models":[52,75,188],"devices":[54,190],"that":[55,90,192],"impedes":[56],"direct":[57],"profiling":[58],"all":[60],"combinations,":[61],"recent":[62],"efforts":[63],"focus":[64],"building":[66],"predictor":[68,167],"to":[69,142,155,168],"different":[77,171,174],"devices.":[78],"However,":[79],"none":[80],"existing":[83],"attempts":[84],"achieved":[86],"cost":[88],"can":[91],"accurately":[92],"predict":[93],"programs":[99],"while":[100],"supporting":[101],"both":[102,118],"training":[103,216],"inference":[105],"accelerators.":[106],"We":[107,123,150],"propose":[108],"CDMPP,":[109],"an":[110,125],"efficient":[111,128],"prediction":[115,202],"framework":[116],"for":[117,165,204],"cross-model":[119,205],"cross-device":[121,207],"prediction.":[122],"design":[124],"informative":[126],"but":[127],"representation":[129],"programs,":[132],"called":[133],"compact":[134],"ASTs,":[135],"pre-order-based":[138],"positional":[139],"encoding":[140],"method,":[141],"capture":[143],"internal":[145],"structure":[146],"programs.":[149],"develop":[151],"domain-adaption-inspired":[153],"method":[154],"learn":[156,169],"domain-invariant":[157],"representations":[158],"devise":[160],"KMeans-based":[162],"sampling":[163],"algorithm,":[164],"from":[170],"domains":[172],"(i.e.,":[173],"operators":[176],"devices).":[178],"Our":[179],"extensive":[180],"experiments":[181],"diverse":[184],"demonstrate":[191],"CDMPP":[193],"significantly":[194],"outperforms":[195],"state-of-the-art":[196],"baselines":[197],"with":[198],"14.03%":[199],"10.85%":[201],"error":[203],"prediction,":[208],"respectively,":[209],"one":[211],"order":[212],"magnitude":[214],"higher":[215],"efficiency.":[217],"The":[218],"implementation":[219],"expanded":[222],"dataset":[223],"are":[224],"available":[225],"at":[226],"https://github.com/joapolarbear/cdmpp.":[227]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}