{"id":"https://openalex.org/W7157647801","doi":"https://doi.org/10.1145/3805621.3807657","title":"DFS: Dynamic Flow Spraying with Bounded Reordering for AI Training Clusters","display_name":"DFS: Dynamic Flow Spraying with Bounded Reordering for AI Training Clusters","publication_year":2026,"publication_date":"2026-04-27","ids":{"openalex":"https://openalex.org/W7157647801","doi":"https://doi.org/10.1145/3805621.3807657"},"language":null,"primary_location":{"id":"doi:10.1145/3805621.3807657","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807657","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805621.3807657","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134827253","display_name":"Guanjun Ma","orcid":"https://orcid.org/0009-0002-1941-1021"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guanjun Ma","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0002-1941-1021","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102593684","display_name":"Chenyue Zheng","orcid":"https://orcid.org/0009-0002-9414-981X"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenyue Zheng","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0002-9414-981X","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5134855634","display_name":"Yuchao Zhang","orcid":"https://orcid.org/0000-0002-0135-8915"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]},{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["CN","GB"],"is_corresponding":false,"raw_author_name":"Yuchao Zhang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China and University of Cambridge, Cambridge, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-0135-8915","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China and University of Cambridge, Cambridge, United Kingdom","institution_ids":["https://openalex.org/I139759216","https://openalex.org/I241749"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"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":0.37353606,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"377","last_page":"385"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10714","display_name":"Software-Defined Networks and 5G","score":0.19910000264644623,"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/T10714","display_name":"Software-Defined Networks and 5G","score":0.19910000264644623,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.08709999918937683,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11896","display_name":"Opportunistic and Delay-Tolerant Networks","score":0.06790000200271606,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6675000190734863},{"id":"https://openalex.org/keywords/network-packet","display_name":"Network packet","score":0.593500018119812},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.517799973487854},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.48969998955726624},{"id":"https://openalex.org/keywords/bounded-function","display_name":"Bounded function","score":0.4848000109195709},{"id":"https://openalex.org/keywords/routing","display_name":"Routing (electronic design automation)","score":0.4652999937534332},{"id":"https://openalex.org/keywords/expert-system","display_name":"Expert system","score":0.46389999985694885},{"id":"https://openalex.org/keywords/hash-function","display_name":"Hash function","score":0.35670000314712524}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.720300018787384},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6675000190734863},{"id":"https://openalex.org/C158379750","wikidata":"https://www.wikidata.org/wiki/Q214111","display_name":"Network packet","level":2,"score":0.593500018119812},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.517799973487854},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.48969998955726624},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.4848000109195709},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.4652999937534332},{"id":"https://openalex.org/C58328972","wikidata":"https://www.wikidata.org/wiki/Q184609","display_name":"Expert system","level":2,"score":0.46389999985694885},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.39399999380111694},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3846000134944916},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.35670000314712524},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.353300005197525},{"id":"https://openalex.org/C114809511","wikidata":"https://www.wikidata.org/wiki/Q1412924","display_name":"Flow network","level":2,"score":0.3458000123500824},{"id":"https://openalex.org/C54108766","wikidata":"https://www.wikidata.org/wiki/Q391064","display_name":"Packet loss","level":3,"score":0.31929999589920044},{"id":"https://openalex.org/C207512268","wikidata":"https://www.wikidata.org/wiki/Q3074551","display_name":"Traffic flow (computer networking)","level":2,"score":0.31189998984336853},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.30410000681877136},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.287200003862381},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28130000829696655},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26170000433921814}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805621.3807657","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807657","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805621.3807657","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807657","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":9,"referenced_works":["https://openalex.org/W1994926493","https://openalex.org/W2169246522","https://openalex.org/W2498764059","https://openalex.org/W2744567246","https://openalex.org/W2968108410","https://openalex.org/W4286308242","https://openalex.org/W4290991080","https://openalex.org/W4394969278","https://openalex.org/W4413756872"],"related_works":[],"abstract_inverted_index":{"Mixture-of-Experts":[0],"(MoE)":[1],"training":[2,15],"in":[3],"modern":[4],"super-node":[5],"AI":[6],"clusters":[7],"is":[8,46],"increasingly":[9],"limited":[10,64],"by":[11],"communication.":[12],"At":[13],"each":[14],"step,":[16],"expert":[17,50,94],"dispatch":[18],"and":[19,41,71,96,126],"combine":[20],"traffic":[21,51,62],"sits":[22],"on":[23],"the":[24,28,69],"critical":[25],"path,":[26],"while":[27],"same":[29],"network":[30],"must":[31],"also":[32],"carry":[33],"throughput-oriented":[34],"transfers":[35],"such":[36],"as":[37],"gradient":[38],"synchronization,":[39],"checkpointing,":[40],"storage":[42],"traffic.":[43],"The":[44],"result":[45],"a":[47,108],"basic":[48],"trade-off:":[49],"benefits":[52],"from":[53],"aggressive":[54],"multi-path":[55],"routing":[56],"to":[57,67,104],"reduce":[58],"latency,":[59],"whereas":[60],"non-expert":[61],"needs":[63],"packet":[65],"reordering":[66],"preserve":[68],"efficiency":[70],"stability":[72],"of":[73],"commodity":[74],"RDMA.":[75],"Existing":[76],"load-balancing":[77],"schemes":[78],"do":[79],"not":[80],"fit":[81],"this":[82],"setting":[83],"well.":[84],"Per-flow":[85],"hashing":[86],"can":[87,120],"create":[88],"short-lived":[89],"hotspots":[90],"that":[91],"delay":[92],"critical-path":[93],"communication,":[95],"reactive":[97],"transport":[98],"mechanisms":[99],"cannot":[100],"respond":[101],"quickly":[102],"enough":[103],"bursty":[105],"dynamics":[106],"within":[107],"single":[109],"RTT.":[110],"Packet":[111],"spraying":[112],"improves":[113],"path":[114],"utilization,":[115],"but":[116],"under":[117],"load":[118],"it":[119],"overflow":[121],"NIC":[122],"out-of-order":[123],"(OOO)":[124],"buffers":[125],"disrupt":[127],"transport.":[128]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-04-30T00:00:00"}