{"id":"https://openalex.org/W7084103366","doi":"https://doi.org/10.1109/infocom55648.2025.11044740","title":"PC4: Precision Collective Communication Congestion Control for AI Cluster","display_name":"PC4: Precision Collective Communication Congestion Control for AI Cluster","publication_year":2025,"publication_date":"2025-05-19","ids":{"openalex":"https://openalex.org/W7084103366","doi":"https://doi.org/10.1109/infocom55648.2025.11044740"},"language":"en","primary_location":{"id":"doi:10.1109/infocom55648.2025.11044740","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom55648.2025.11044740","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2025 - IEEE Conference on Computer Communications","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":null,"display_name":"Taoran Qi","orcid":null},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Taoran Qi","raw_affiliation_strings":["School of Intergrated Circuit Science and Engineering, Beihang University,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Intergrated Circuit Science and Engineering, Beihang University,Beijing,China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Shuo Li","orcid":null},"institutions":[{"id":"https://openalex.org/I4210097127","display_name":"Advanced Interconnect Technologies (China)","ror":"https://ror.org/00zgs1h87","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210097127"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuo Li","raw_affiliation_strings":["Shanghai Enflame Technology Company Limited,Department of Interconnect,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Enflame Technology Company Limited,Department of Interconnect,China","institution_ids":["https://openalex.org/I4210097127"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xingqi Zou","orcid":null},"institutions":[{"id":"https://openalex.org/I4210097127","display_name":"Advanced Interconnect Technologies (China)","ror":"https://ror.org/00zgs1h87","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210097127"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingqi Zou","raw_affiliation_strings":["Shanghai Enflame Technology Company Limited,Department of Interconnect,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Enflame Technology Company Limited,Department of Interconnect,China","institution_ids":["https://openalex.org/I4210097127"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Liangce Deng","orcid":null},"institutions":[{"id":"https://openalex.org/I4210097127","display_name":"Advanced Interconnect Technologies (China)","ror":"https://ror.org/00zgs1h87","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210097127"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liangce Deng","raw_affiliation_strings":["Shanghai Enflame Technology Company Limited,Department of Interconnect,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Enflame Technology Company Limited,Department of Interconnect,China","institution_ids":["https://openalex.org/I4210097127"]}]},{"author_position":"last","author":{"id":null,"display_name":"Guodong Wei","orcid":null},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guodong Wei","raw_affiliation_strings":["School of Intergrated Circuit Science and Engineering, Beihang University,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Intergrated Circuit Science and Engineering, Beihang University,Beijing,China","institution_ids":["https://openalex.org/I82880672"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.7335,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.79932199,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10163","display_name":"Metal complexes synthesis and properties","score":0.42660000920295715,"subfield":{"id":"https://openalex.org/subfields/2730","display_name":"Oncology"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T10163","display_name":"Metal complexes synthesis and properties","score":0.42660000920295715,"subfield":{"id":"https://openalex.org/subfields/2730","display_name":"Oncology"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10464","display_name":"Crystal structures of chemical compounds","score":0.3172999918460846,"subfield":{"id":"https://openalex.org/subfields/1604","display_name":"Inorganic Chemistry"},"field":{"id":"https://openalex.org/fields/16","display_name":"Chemistry"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10096","display_name":"Metal-Organic Frameworks: Synthesis and Applications","score":0.06080000102519989,"subfield":{"id":"https://openalex.org/subfields/1604","display_name":"Inorganic Chemistry"},"field":{"id":"https://openalex.org/fields/16","display_name":"Chemistry"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/network-congestion","display_name":"Network congestion","score":0.6561999917030334},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.5485000014305115},{"id":"https://openalex.org/keywords/queue","display_name":"Queue","score":0.5446000099182129},{"id":"https://openalex.org/keywords/queueing-theory","display_name":"Queueing theory","score":0.5322999954223633},{"id":"https://openalex.org/keywords/network-topology","display_name":"Network topology","score":0.4763999879360199},{"id":"https://openalex.org/keywords/cluster","display_name":"Cluster (spacecraft)","score":0.45249998569488525},{"id":"https://openalex.org/keywords/telecommunications-network","display_name":"Telecommunications network","score":0.4465999901294708},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.4275999963283539},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.42179998755455017}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7555000185966492},{"id":"https://openalex.org/C195563490","wikidata":"https://www.wikidata.org/wiki/Q180368","display_name":"Network congestion","level":3,"score":0.6561999917030334},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6060000061988831},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.5485000014305115},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.5446000099182129},{"id":"https://openalex.org/C22684755","wikidata":"https://www.wikidata.org/wiki/Q847526","display_name":"Queueing theory","level":2,"score":0.5322999954223633},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.4878000020980835},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.4763999879360199},{"id":"https://openalex.org/C164866538","wikidata":"https://www.wikidata.org/wiki/Q367351","display_name":"Cluster (spacecraft)","level":2,"score":0.45249998569488525},{"id":"https://openalex.org/C192126672","wikidata":"https://www.wikidata.org/wiki/Q1068715","display_name":"Telecommunications network","level":2,"score":0.4465999901294708},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.4275999963283539},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.42179998755455017},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.398499995470047},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.3856000006198883},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.375900000333786},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3700000047683716},{"id":"https://openalex.org/C68649174","wikidata":"https://www.wikidata.org/wiki/Q1379116","display_name":"Base station","level":2,"score":0.36640000343322754},{"id":"https://openalex.org/C2988430800","wikidata":"https://www.wikidata.org/wiki/Q428971","display_name":"Scaling law","level":3,"score":0.3587999939918518},{"id":"https://openalex.org/C101765175","wikidata":"https://www.wikidata.org/wiki/Q577764","display_name":"Communications system","level":2,"score":0.35199999809265137},{"id":"https://openalex.org/C2779888511","wikidata":"https://www.wikidata.org/wiki/Q244156","display_name":"Traffic congestion","level":2,"score":0.35109999775886536},{"id":"https://openalex.org/C29140674","wikidata":"https://www.wikidata.org/wiki/Q206637","display_name":"Computer cluster","level":2,"score":0.33629998564720154},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.3321000039577484},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.31150001287460327},{"id":"https://openalex.org/C100339178","wikidata":"https://www.wikidata.org/wiki/Q2548752","display_name":"Collective behavior","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.290800005197525},{"id":"https://openalex.org/C170122806","wikidata":"https://www.wikidata.org/wiki/Q1914828","display_name":"Linear scale","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.2587999999523163}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/infocom55648.2025.11044740","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom55648.2025.11044740","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2025 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"evolution":[1],"of":[2,22,24,35,45,80,160],"AI":[3,81,95],"has":[4],"driven":[5],"the":[6,31,36,43,74],"trend":[7],"towards":[8],"utilizing":[9,147],"larger":[10],"models":[11],"with":[12,42],"increasing":[13],"parameter":[14],"sizes,":[15],"necessitating":[16],"deployment":[17],"in":[18,110,118,180],"datacenters":[19],"comprising":[20],"tens":[21],"thousands":[23],"GPUs.":[25],"However,":[26],"as":[27,150],"cluster":[28],"sizes":[29],"expand,":[30],"overall":[32],"computational":[33],"capacity":[34],"system":[37,137],"fails":[38],"to":[39,55,184],"scale":[40],"linearly":[41],"number":[44],"compute":[46,130],"nodes.":[47],"This":[48,153],"non-linear":[49],"scaling":[50],"law":[51],"is":[52],"primarily":[53],"due":[54],"decreased":[56],"GPU":[57],"utilization":[58],"caused":[59],"by":[60],"network":[61,75],"congestion,":[62],"which":[63],"significantly":[64],"impacts":[65],"model":[66],"training":[67],"speed.":[68],"In":[69,171],"this":[70],"paper,":[71],"we":[72,129],"analyze":[73],"topology":[76],"and":[77,83,164],"traffic":[78],"characteristics":[79],"clusters":[82],"propose":[84],"PC4,":[85],"a":[86,131,142,177],"precision":[87],"collective":[88,103,111,126],"communication":[89,104,112,127],"congestion":[90,99,151],"control":[91,100,159],"algorithm":[92],"tailored":[93],"for":[94,156],"clusters.":[96],"PC4":[97,140,175],"optimizes":[98],"based":[101],"on":[102],"operations":[105],"(e.g.,":[106,114],"all-reduce,":[107],"all-to-all)":[108],"inherent":[109],"libraries":[113],"MPI,":[115],"NCCL)":[116],"used":[117],"distributed":[119],"training.":[120],"By":[121],"leveraging":[122],"information":[123],"from":[124],"these":[125],"operations,":[128],"base":[132],"rate":[133],"that":[134],"enables":[135],"rapid":[136],"convergence.":[138],"Furthermore,":[139],"employs":[141],"datacenter":[143],"time":[144],"synchronization":[145],"mechanism,":[146],"one-way":[148],"delay":[149],"signal.":[152],"approach":[154],"allows":[155],"more":[157],"precise":[158],"link":[161],"queue":[162],"lengths":[163],"queuing":[165],"time,":[166],"thereby":[167],"efficiently":[168],"managing":[169],"congestion.":[170],"our":[172],"all-to-all":[173],"evaluation,":[174],"achieves":[176],"72%":[178],"reduction":[179],"tail":[181],"FCT":[182],"compared":[183],"DCQCN.":[185]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}
