{"id":"https://openalex.org/W7084062718","doi":"https://doi.org/10.1109/icc52391.2025.11160830","title":"FlowTracer: A Tool for Uncovering Network Path Usage Imbalance in AI Training Clusters","display_name":"FlowTracer: A Tool for Uncovering Network Path Usage Imbalance in AI Training Clusters","publication_year":2025,"publication_date":"2025-06-08","ids":{"openalex":"https://openalex.org/W7084062718","doi":"https://doi.org/10.1109/icc52391.2025.11160830"},"language":"en","primary_location":{"id":"doi:10.1109/icc52391.2025.11160830","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icc52391.2025.11160830","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICC 2025 - IEEE International Conference on 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":"Hasibul Jamil","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hasibul Jamil","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Abdul Alim","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abdul Alim","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Laurent Schares","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Laurent Schares","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Pavlos Maniotis","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pavlos Maniotis","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Liran Schour","orcid":null},"institutions":[{"id":"https://openalex.org/I4210167297","display_name":"IBM Research - Haifa","ror":"https://ror.org/05rw9t746","country_code":"IL","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210167297"]}],"countries":["IL"],"is_corresponding":false,"raw_author_name":"Liran Schour","raw_affiliation_strings":["IBM Research Lab,Haifa,Israel"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research Lab,Haifa,Israel","institution_ids":["https://openalex.org/I4210167297"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ali Sydney","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ali Sydney","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Abdullah Kayi","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abdullah Kayi","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tevfik Kosar","orcid":null},"institutions":[{"id":"https://openalex.org/I63190737","display_name":"University at Buffalo, State University of New York","ror":"https://ror.org/01y64my43","country_code":"US","type":"education","lineage":["https://openalex.org/I63190737"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tevfik Kosar","raw_affiliation_strings":["University at Buffalo (SUNY),Buffalo,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University at Buffalo (SUNY),Buffalo,NY,USA","institution_ids":["https://openalex.org/I63190737"]}]},{"author_position":"last","author":{"id":null,"display_name":"Bengi Karacali","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bengi Karacali","raw_affiliation_strings":["IBM Research,NY,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,NY,USA","institution_ids":["https://openalex.org/I1341412227"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":9,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.50275443,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6988","last_page":"6993"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10506","display_name":"Aquaculture disease management and microbiota","score":0.25440001487731934,"subfield":{"id":"https://openalex.org/subfields/2403","display_name":"Immunology"},"field":{"id":"https://openalex.org/fields/24","display_name":"Immunology and Microbiology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T10506","display_name":"Aquaculture disease management and microbiota","score":0.25440001487731934,"subfield":{"id":"https://openalex.org/subfields/2403","display_name":"Immunology"},"field":{"id":"https://openalex.org/fields/24","display_name":"Immunology and Microbiology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12824","display_name":"Infections and bacterial resistance","score":0.1931000053882599,"subfield":{"id":"https://openalex.org/subfields/1310","display_name":"Endocrinology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T13350","display_name":"Myxozoan Parasites in Aquatic Species","score":0.04010000079870224,"subfield":{"id":"https://openalex.org/subfields/1306","display_name":"Cancer Research"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.5547000169754028},{"id":"https://openalex.org/keywords/debugging","display_name":"Debugging","score":0.5328999757766724},{"id":"https://openalex.org/keywords/routing","display_name":"Routing (electronic design automation)","score":0.526199996471405},{"id":"https://openalex.org/keywords/routing-domain","display_name":"Routing domain","score":0.4535999894142151},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.42590001225471497},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4221999943256378},{"id":"https://openalex.org/keywords/static-routing","display_name":"Static routing","score":0.4187999963760376},{"id":"https://openalex.org/keywords/hash-function","display_name":"Hash function","score":0.38940000534057617},{"id":"https://openalex.org/keywords/routing-protocol","display_name":"Routing protocol","score":0.37869998812675476}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.82669997215271},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6590999960899353},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.5547000169754028},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.5333999991416931},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.5328999757766724},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.526199996471405},{"id":"https://openalex.org/C94600068","wikidata":"https://www.wikidata.org/wiki/Q7371621","display_name":"Routing domain","level":5,"score":0.4535999894142151},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.42590001225471497},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4221999943256378},{"id":"https://openalex.org/C204948658","wikidata":"https://www.wikidata.org/wiki/Q1119410","display_name":"Static routing","level":4,"score":0.4187999963760376},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.38940000534057617},{"id":"https://openalex.org/C104954878","wikidata":"https://www.wikidata.org/wiki/Q1648707","display_name":"Routing protocol","level":3,"score":0.37869998812675476},{"id":"https://openalex.org/C184896649","wikidata":"https://www.wikidata.org/wiki/Q290066","display_name":"Routing table","level":4,"score":0.37549999356269836},{"id":"https://openalex.org/C196423136","wikidata":"https://www.wikidata.org/wiki/Q7209671","display_name":"Policy-based routing","level":5,"score":0.37439998984336853},{"id":"https://openalex.org/C67388219","wikidata":"https://www.wikidata.org/wiki/Q207440","display_name":"Hash table","level":3,"score":0.33480000495910645},{"id":"https://openalex.org/C76522221","wikidata":"https://www.wikidata.org/wiki/Q5035396","display_name":"Multipath routing","level":5,"score":0.32100000977516174},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2890999913215637},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.2773999869823456},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.2768999934196472},{"id":"https://openalex.org/C195780805","wikidata":"https://www.wikidata.org/wiki/Q1535986","display_name":"Metrics","level":5,"score":0.27379998564720154},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C89305328","wikidata":"https://www.wikidata.org/wiki/Q1755411","display_name":"Link-state routing protocol","level":4,"score":0.2606000006198883},{"id":"https://openalex.org/C114809511","wikidata":"https://www.wikidata.org/wiki/Q1412924","display_name":"Flow network","level":2,"score":0.25360000133514404},{"id":"https://openalex.org/C133474645","wikidata":"https://www.wikidata.org/wiki/Q17104660","display_name":"Network mapping","level":3,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icc52391.2025.11160830","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icc52391.2025.11160830","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICC 2025 - IEEE International Conference on Communications","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1509372741","display_name":null,"funder_award_id":"OAC-2313061","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,190],"increasing":[1],"complexity":[2],"of":[3,20,131,153,182],"AI":[4,155],"workloads,":[5],"especially":[6],"distributed":[7,154],"Large":[8],"Language":[9],"Model":[10],"(LLM)":[11],"training,":[12],"places":[13],"significant":[14],"strain":[15],"on":[16],"the":[17,128,151,179],"networking":[18],"infrastructure":[19],"parallel":[21,35],"data":[22,95],"centers":[23],"and":[24,46,61,81,109,124,149,166],"supercomputing":[25],"systems.":[26,115],"While":[27],"Equal-Cost":[28],"Multi-Path":[29],"(ECMP)":[30],"routing":[31,64,111,184],"distributes":[32],"traffic":[33,122],"over":[34],"paths,":[36],"hash":[37,136],"collisions":[38],"often":[39],"lead":[40],"to":[41,56,170,177],"imbalanced":[42],"network":[43,58,76,165],"resource":[44],"utilization":[45,60],"performance":[47,132,152],"bottlenecks.":[48],"This":[49],"paper":[50],"presents":[51],"FlowTracer,":[52],"a":[53,87,159,163,186,193,201],"tool":[54],"designed":[55],"analyze":[57],"path":[59],"evaluate":[62],"different":[63],"strategies.":[65],"Unlike":[66],"tools":[67],"that":[68],"introduce":[69],"additional":[70],"traffic,":[71],"FlowTracer":[72,89,116,173],"aids":[73],"in":[74,113,196],"debugging":[75,108],"inefficiencies":[77],"by":[78,200],"passively":[79],"monitoring":[80],"correlating":[82],"user":[83],"workload":[84],"flows.":[85],"As":[86],"result,":[88],"does":[90],"not":[91],"interfere":[92],"with":[93,99,162],"ongoing":[94],"transfers,":[96],"enabling":[97],"analysis":[98],"minimal":[100],"overhead,":[101],"which":[102],"is":[103],"an":[104],"important":[105],"factor":[106],"when":[107],"fine-tuning":[110],"schemes":[112],"production":[114],"can":[117,125,144,174],"provide":[118],"detailed":[119],"insights":[120],"into":[121],"distribution":[123],"help":[126],"identify":[127],"root":[129],"causes":[130],"degradation,":[133],"such":[134],"as":[135,198],"collisions.":[137],"With":[138],"FlowTracer's":[139],"flow-level":[140],"insights,":[141],"system":[142],"operators":[143],"optimize":[145],"routing,":[146],"reduce":[147],"congestion,":[148],"improve":[150],"workloads.":[156],"We":[157],"use":[158],"RoCEv2-enabled":[160],"cluster":[161],"leaf-spine":[164],"16":[167],"400-Gbps":[168],"nodes":[169],"demonstrate":[171],"how":[172],"be":[175],"used":[176],"compare":[178],"flow":[180],"imbalances":[181],"ECMP":[183],"against":[185],"statically":[187],"configured":[188],"network.":[189],"example":[191],"showcases":[192],"30%":[194],"reduction":[195],"imbalance,":[197],"measured":[199],"new":[202],"metric":[203],"we":[204],"introduce.":[205]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}