{"id":"https://openalex.org/W4415124497","doi":"https://doi.org/10.1109/icnp65844.2025.11192456","title":"LIBRA: Multi-Path Transport in Distributed Training Data Center Network","display_name":"LIBRA: Multi-Path Transport in Distributed Training Data Center Network","publication_year":2025,"publication_date":"2025-09-22","ids":{"openalex":"https://openalex.org/W4415124497","doi":"https://doi.org/10.1109/icnp65844.2025.11192456"},"language":"en","primary_location":{"id":"doi:10.1109/icnp65844.2025.11192456","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icnp65844.2025.11192456","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 33rd International Conference on Network Protocols (ICNP)","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/A5102610881","display_name":"Xinshuo Wang","orcid":"https://orcid.org/0009-0006-3658-779X"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinshuo Wang","raw_affiliation_strings":["Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102011601","display_name":"Baihua Chen","orcid":"https://orcid.org/0000-0002-0179-6288"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Baihua Chen","raw_affiliation_strings":["Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics","institution_ids":["https://openalex.org/I4210099069"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029035526","display_name":"Lei Liu","orcid":"https://orcid.org/0000-0001-5579-5419"},"institutions":[{"id":"https://openalex.org/I4210099069","display_name":"Institute of Acoustics","ror":"https://ror.org/00v8rqv75","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210099069"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Liu","raw_affiliation_strings":["Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,National Network New Media Engineering Research Center, Institute of Acoustics","institution_ids":["https://openalex.org/I4210099069"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210099069"],"apc_list":null,"apc_paid":null,"fwci":1.8777,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.88700483,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"12"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.7814000248908997,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.7814000248908997,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.7664999961853027,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T12292","display_name":"Graph Theory and Algorithms","score":0.724399983882904,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/remote-direct-memory-access","display_name":"Remote direct memory access","score":0.7587000131607056},{"id":"https://openalex.org/keywords/network-congestion","display_name":"Network congestion","score":0.5078999996185303},{"id":"https://openalex.org/keywords/data-center","display_name":"Data center","score":0.45159998536109924},{"id":"https://openalex.org/keywords/data-transmission","display_name":"Data transmission","score":0.44609999656677246},{"id":"https://openalex.org/keywords/load-balancing","display_name":"Load balancing (electrical power)","score":0.42809998989105225},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.3736000061035156},{"id":"https://openalex.org/keywords/network-interface","display_name":"Network interface","score":0.3614000082015991},{"id":"https://openalex.org/keywords/infiniband","display_name":"InfiniBand","score":0.3610000014305115},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.35749998688697815},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.3407000005245209}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8184000253677368},{"id":"https://openalex.org/C130795937","wikidata":"https://www.wikidata.org/wiki/Q2561570","display_name":"Remote direct memory access","level":2,"score":0.7587000131607056},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.614300012588501},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.5929999947547913},{"id":"https://openalex.org/C195563490","wikidata":"https://www.wikidata.org/wiki/Q180368","display_name":"Network congestion","level":3,"score":0.5078999996185303},{"id":"https://openalex.org/C153740404","wikidata":"https://www.wikidata.org/wiki/Q671224","display_name":"Data center","level":2,"score":0.45159998536109924},{"id":"https://openalex.org/C557945733","wikidata":"https://www.wikidata.org/wiki/Q389772","display_name":"Data transmission","level":2,"score":0.44609999656677246},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.42809998989105225},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.3736000061035156},{"id":"https://openalex.org/C103987645","wikidata":"https://www.wikidata.org/wiki/Q985806","display_name":"Network interface","level":3,"score":0.3614000082015991},{"id":"https://openalex.org/C2781030343","wikidata":"https://www.wikidata.org/wiki/Q922437","display_name":"InfiniBand","level":2,"score":0.3610000014305115},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.35749998688697815},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.3407000005245209},{"id":"https://openalex.org/C761482","wikidata":"https://www.wikidata.org/wiki/Q118093","display_name":"Transmission (telecommunications)","level":2,"score":0.3352000117301941},{"id":"https://openalex.org/C159631557","wikidata":"https://www.wikidata.org/wiki/Q1546066","display_name":"Networking hardware","level":2,"score":0.33410000801086426},{"id":"https://openalex.org/C113843644","wikidata":"https://www.wikidata.org/wiki/Q901882","display_name":"Interface (matter)","level":4,"score":0.3287999927997589},{"id":"https://openalex.org/C22029948","wikidata":"https://www.wikidata.org/wiki/Q45089","display_name":"Dice","level":2,"score":0.32739999890327454},{"id":"https://openalex.org/C139940560","wikidata":"https://www.wikidata.org/wiki/Q290036","display_name":"Network simulation","level":2,"score":0.3257000148296356},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.321399986743927},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C203274722","wikidata":"https://www.wikidata.org/wiki/Q7001161","display_name":"Network performance","level":2,"score":0.30329999327659607},{"id":"https://openalex.org/C193519340","wikidata":"https://www.wikidata.org/wiki/Q891179","display_name":"Data loss","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C171659815","wikidata":"https://www.wikidata.org/wiki/Q165233","display_name":"Network interface controller","level":2,"score":0.2992999851703644},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.27630001306533813},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C184896649","wikidata":"https://www.wikidata.org/wiki/Q290066","display_name":"Routing table","level":4,"score":0.25769999623298645},{"id":"https://openalex.org/C77270119","wikidata":"https://www.wikidata.org/wiki/Q1655198","display_name":"Software-defined networking","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2565000057220459},{"id":"https://openalex.org/C201100257","wikidata":"https://www.wikidata.org/wiki/Q393287","display_name":"Network traffic control","level":3,"score":0.25589999556541443},{"id":"https://openalex.org/C70061542","wikidata":"https://www.wikidata.org/wiki/Q989016","display_name":"Distributed database","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.25270000100135803},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icnp65844.2025.11192456","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icnp65844.2025.11192456","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 33rd International Conference on Network Protocols (ICNP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W1968331080","https://openalex.org/W2036003010","https://openalex.org/W2038718790","https://openalex.org/W2130531694","https://openalex.org/W2142480021","https://openalex.org/W2158733823","https://openalex.org/W2292669715","https://openalex.org/W2498764059","https://openalex.org/W2968108410","https://openalex.org/W2982686855","https://openalex.org/W2985042979","https://openalex.org/W3034977853","https://openalex.org/W3049093198","https://openalex.org/W3101455050","https://openalex.org/W3175449831","https://openalex.org/W4237515752","https://openalex.org/W4290991674","https://openalex.org/W4401176521","https://openalex.org/W4401176799"],"related_works":[],"abstract_inverted_index":{"Effective":[0],"networking":[1],"is":[2],"crucial":[3],"for":[4,91,122,170,203],"the":[5,49,61,69,120,167,189,218],"efficient":[6,98],"and":[7,31,129,154,177,186],"cost-effective":[8],"training":[9,57],"of":[10,52,64,71,100,166,220],"artificial":[11],"intelligence":[12],"(AI)":[13],"models.":[14],"Remote":[15],"Direct":[16],"Memory":[17],"Access":[18],"(RDMA)":[19],"has":[20],"gained":[21],"widespread":[22],"adoption":[23],"due":[24,75],"to":[25,76,79,139],"its":[26],"low":[27],"latency,":[28],"high":[29],"throughput,":[30],"minimal":[32],"CPU":[33],"overhead.":[34],"However,":[35],"current":[36],"RDMA":[37,92],"implementations":[38],"remain":[39],"constrained":[40],"by":[41],"underlying":[42],"Equal-Cost":[43],"Multi-Path":[44],"(ECMP)":[45],"forwarding":[46],"mechanisms,":[47],"hindering":[48],"full":[50],"exploitation":[51],"parallel":[53],"paths":[54,103],"within":[55,93,193],"distributed":[56],"data":[58,106,194],"centers.":[59,107],"Furthermore,":[60,134],"rapid":[62],"evolution":[63],"network":[65,81,95,102,151,191],"interface":[66],"bandwidth":[67],"necessitates":[68],"re-evaluation":[70],"existing":[72],"multi-path":[73,88,114,141],"approaches":[74],"ineffective":[77],"scaling":[78],"high-speed":[80,94],"environments.This":[82],"study":[83],"introduces":[84],"LIBRA,":[85],"a":[86,197],"novel":[87,135],"transmission":[89,206],"method":[90],"environments,":[96],"enabling":[97],"utilization":[99],"abundant":[101],"in":[104,200],"modern":[105],"LIBRA":[108,171,184],"integrates":[109],"segment":[110],"routing":[111],"with":[112],"packet-level":[113],"spraying.":[115],"Deterministic":[116],"path":[117,124],"assignment":[118],"eliminates":[119],"need":[121],"sender-side":[123],"detection,":[125],"simplifying":[126],"load-balancing":[127],"design":[128],"enhancing":[130],"overall":[131],"system":[132],"efficiency.":[133],"algorithms":[136,149],"are":[137],"proposed":[138],"decouple":[140],"load":[142],"balancing":[143],"from":[144],"congestion":[145,152],"control":[146,158,162],"(CC).":[147],"These":[148],"identify":[150],"positions":[153],"implement":[155],"tailored":[156],"traffic":[157],"strategies":[159],"through":[160],"fine-grained":[161],"mechanisms.":[163],"The":[164],"feasibility":[165],"hardware":[168],"implementation":[169],"was":[172],"experimentally":[173],"validated":[174],"using":[175],"FPGA":[176],"Tofino-based":[178],"prototypes.":[179],"Large-scale":[180],"simulations":[181],"demonstrate":[182],"that":[183],"effectively":[185],"robustly":[187],"leverages":[188],"rich":[190],"multi-paths":[192],"centers,":[195],"yielding":[196],"2~4\u00d7":[198],"reduction":[199],"tail":[201],"latency":[202],"large":[204],"message":[205],"scenarios.":[207],"Our":[208],"simulation":[209],"code":[210],"will":[211],"be":[212],"released":[213],"as":[214],"open":[215],"source":[216],"at":[217],"time":[219],"publication.":[221]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-14T00:00:00"}
