{"id":"https://openalex.org/W4392182051","doi":"https://doi.org/10.1145/3637528.3671844","title":"MSPipe: Efficient Temporal GNN Training via Staleness-Aware Pipeline","display_name":"MSPipe: Efficient Temporal GNN Training via Staleness-Aware Pipeline","publication_year":2024,"publication_date":"2024-08-24","ids":{"openalex":"https://openalex.org/W4392182051","doi":"https://doi.org/10.1145/3637528.3671844"},"language":"en","primary_location":{"id":"doi:10.1145/3637528.3671844","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3637528.3671844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2402.15113","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020661868","display_name":"Guangming Sheng","orcid":"https://orcid.org/0000-0003-3395-3994"},"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":"Guangming Sheng","raw_affiliation_strings":["The University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-3395-3994","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":null,"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":["The University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-0537-4004","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chao Huang","orcid":"https://orcid.org/0000-0002-2062-1512"},"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":"Chao Huang","raw_affiliation_strings":["The University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-2062-1512","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101232348","display_name":"Chuan Wu","orcid":null},"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":["The University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-3144-4398","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":1.974,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.8735289,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"2651","last_page":"2662"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9912999868392944,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9912999868392944,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9635999798774719,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9437000155448914,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.8320802450180054},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6948482990264893},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6142333149909973},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.06977000832557678},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.06866958737373352},{"id":"https://openalex.org/keywords/meteorology","display_name":"Meteorology","score":0.05716624855995178}],"concepts":[{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.8320802450180054},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6948482990264893},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6142333149909973},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.06977000832557678},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.06866958737373352},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.05716624855995178}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3637528.3671844","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3637528.3671844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2402.15113","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.15113","pdf_url":"https://arxiv.org/pdf/2402.15113","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2402.15113","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.15113","pdf_url":"https://arxiv.org/pdf/2402.15113","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.5899999737739563,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2124738084","https://openalex.org/W2798918712","https://openalex.org/W2807021761","https://openalex.org/W2963076818","https://openalex.org/W2965683718","https://openalex.org/W2966573219","https://openalex.org/W2972577885","https://openalex.org/W2998116985","https://openalex.org/W3012588100","https://openalex.org/W3024560045","https://openalex.org/W3035178789","https://openalex.org/W3043590771","https://openalex.org/W3080512189","https://openalex.org/W3093933627","https://openalex.org/W3109841242","https://openalex.org/W3206830733","https://openalex.org/W4220807331","https://openalex.org/W4283314525","https://openalex.org/W4288070868","https://openalex.org/W4321445982"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W2358668433","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2997094352"],"abstract_inverted_index":{"Memory-based":[0],"Temporal":[1],"Graph":[2],"Neural":[3],"Networks":[4],"(MTGNNs)":[5],"are":[6,71],"a":[7,16,89,117,160,231,276],"class":[8],"of":[9,42,88,179,285],"temporal":[10,25,56,103,170,201],"graph":[11],"neural":[12],"networks":[13],"that":[14,125,198,249,264],"utilize":[15],"node":[17,145],"memory":[18,44,90,146,155,166,208,213],"module":[19,45,167],"to":[20,28,32,48,53,75,78,168,174,210,235,268],"capture":[21],"and":[22,39,62,85,119,143,177,184,206,239,247],"retain":[23],"long-term":[24],"dependencies,":[26,104],"leading":[27],"superior":[29],"performance":[30,176],"compared":[31],"memory-less":[33],"counterparts.":[34],"However,":[35,157],"the":[36,43,55,86,99,137,154,165,220,252,292],"iterative":[37],"reading":[38],"updating":[40,144],"process":[41],"in":[46,80,148,164,196],"MTGNNs":[47,76,149],"obtain":[49,211],"up-to-date":[50],"information":[51],"needs":[52],"follow":[54],"dependencies.":[57],"This":[58,215],"introduces":[59],"significant":[60],"overhead":[61],"limits":[63],"training":[64,81,127,222,237],"throughput.":[65],"Existing":[66],"optimizations":[67,94],"for":[68,108,122,279],"static":[69],"GNNs":[70],"not":[72,96],"directly":[73],"applicable":[74],"due":[77],"differences":[79],"paradigm,":[82],"model":[83,131,240],"architecture,":[84],"absence":[87],"module.":[91,156],"Moreover,":[92],"these":[93],"do":[95],"effectively":[97],"address":[98],"challenges":[100,139],"posed":[101],"by":[102,150],"making":[105,274],"them":[106],"ineffective":[107],"MTGNN":[109,221,281],"training.":[110,260,282],"In":[111],"this":[112],"paper,":[113],"we":[114,189,229,243],"propose":[115],"MSPipe,":[116],"general":[118],"efficient":[120,280],"framework":[121],"memory-based":[123],"TGNNs":[124],"maximizes":[126],"throughput":[128],"while":[129],"maintaining":[130],"accuracy.":[132,241],"Our":[133],"design":[134,230],"specifically":[135],"addresses":[136],"unique":[138],"associated":[140],"with":[141,203],"fetching":[142,209],"states":[147],"integrating":[151],"staleness":[152,162,205,232],"into":[153],"simply":[158],"introducing":[159],"predefined":[161],"bound":[163],"break":[169],"dependencies":[171,202],"may":[172],"lead":[173],"suboptimal":[175],"lack":[178],"generalizability":[180],"across":[181],"different":[182],"models":[183],"datasets.":[185],"To":[186],"overcome":[187],"this,":[188],"introduce":[190],"an":[191],"online":[192],"pipeline":[193],"scheduling":[194],"algorithm":[195],"MSPipe":[197,250,265],"strategically":[199],"breaks":[200],"minimal":[204],"delays":[207],"fresher":[212],"states.":[214],"is":[216],"achieved":[217],"without":[218,271],"stalling":[219],"stage":[223],"or":[224],"causing":[225],"resource":[226],"contention.":[227],"Additionally,":[228],"mitigation":[233],"mechanism":[234],"enhance":[236],"convergence":[238,245,254],"Furthermore,":[242],"provide":[244],"analysis":[246],"demonstrate":[248],"maintains":[251],"same":[253],"rate":[255],"as":[256],"vanilla":[257],"sampling-based":[258],"GNN":[259],"Experimental":[261],"results":[262],"show":[263],"achieves":[266],"up":[267],"2.45\u00d7":[269],"speed-up":[270],"sacrificing":[272],"accuracy,":[273],"it":[275],"promising":[277],"solution":[278],"The":[283],"implementation":[284],"our":[286],"paper":[287],"can":[288],"be":[289],"found":[290],"at":[291],"following":[293],"link:":[294],"https://github.com/PeterSH6/MSPipe.":[295]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}