{"id":"https://openalex.org/W4403794376","doi":"https://doi.org/10.1093/comjnl/bxae110","title":"A resource-aware workload scheduling method for unbalanced GEMMs on GPUs","display_name":"A resource-aware workload scheduling method for unbalanced GEMMs on GPUs","publication_year":2024,"publication_date":"2024-10-25","ids":{"openalex":"https://openalex.org/W4403794376","doi":"https://doi.org/10.1093/comjnl/bxae110"},"language":"en","primary_location":{"id":"doi:10.1093/comjnl/bxae110","is_oa":false,"landing_page_url":"https://doi.org/10.1093/comjnl/bxae110","pdf_url":null,"source":{"id":"https://openalex.org/S44643521","display_name":"The Computer Journal","issn_l":"0010-4620","issn":["0010-4620","1460-2067"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310311648","host_organization_name":"Oxford University Press","host_organization_lineage":["https://openalex.org/P4310311648","https://openalex.org/P4310311647"],"host_organization_lineage_names":["Oxford University Press","University of Oxford"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Computer Journal","raw_type":"journal-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/A5104298138","display_name":"Hui Liu","orcid":"https://orcid.org/0000-0002-7087-2416"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Hangda Liu","raw_affiliation_strings":["Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","University of Chinese Academy of Sciences , Beijing 100190,","Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China","University of Chinese Academy of Sciences, Beijing 100190, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences , Beijing 100190,","institution_ids":["https://openalex.org/I4210165038"]},{"raw_affiliation_string":"Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060260820","display_name":"Boyu Diao","orcid":null},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Boyu Diao","raw_affiliation_strings":["Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]},{"raw_affiliation_string":"Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014698348","display_name":"Wenxing Chen","orcid":"https://orcid.org/0000-0001-9669-4358"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenxin Chen","raw_affiliation_strings":["Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","University of Chinese Academy of Sciences , Beijing 100190,","University of Chinese Academy of Sciences, Beijing 100190, China","Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences , Beijing 100190,","institution_ids":["https://openalex.org/I4210165038"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210165038"]},{"raw_affiliation_string":"Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103245119","display_name":"Yongjun Xu","orcid":"https://orcid.org/0000-0001-6647-0986"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongjun Xu","raw_affiliation_strings":["Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology , Chinese Academy of Sciences, Beijing 100190,","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]},{"raw_affiliation_string":"Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5104298138"],"corresponding_institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210090176","https://openalex.org/I4210165038"],"apc_list":{"value":2635,"currency":"GBP","value_usd":3232},"apc_paid":null,"fwci":2.7482,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.91955738,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"68","issue":"3","first_page":"273","last_page":"282"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9993000030517578,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9932000041007996,"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/computer-science","display_name":"Computer science","score":0.8370606899261475},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6867940425872803},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6263766288757324},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5114625692367554},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3393021523952484},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.3145020604133606},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.08210772275924683},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.04837962985038757}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8370606899261475},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6867940425872803},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6263766288757324},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5114625692367554},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3393021523952484},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.3145020604133606},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.08210772275924683},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.04837962985038757}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1093/comjnl/bxae110","is_oa":false,"landing_page_url":"https://doi.org/10.1093/comjnl/bxae110","pdf_url":null,"source":{"id":"https://openalex.org/S44643521","display_name":"The Computer Journal","issn_l":"0010-4620","issn":["0010-4620","1460-2067"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310311648","host_organization_name":"Oxford University Press","host_organization_lineage":["https://openalex.org/P4310311648","https://openalex.org/P4310311647"],"host_organization_lineage_names":["Oxford University Press","University of Oxford"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Computer Journal","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8","score":0.4699999988079071}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2001804590","https://openalex.org/W2149381887","https://openalex.org/W2499931820","https://openalex.org/W3043065011","https://openalex.org/W3095319910","https://openalex.org/W3176305871","https://openalex.org/W3210165781","https://openalex.org/W4213019189","https://openalex.org/W4221037885","https://openalex.org/W4280633903","https://openalex.org/W4294791686","https://openalex.org/W4295312788","https://openalex.org/W4321052080","https://openalex.org/W4385778234","https://openalex.org/W4393159688","https://openalex.org/W6739901393","https://openalex.org/W6745021780","https://openalex.org/W6751349269","https://openalex.org/W6766978945","https://openalex.org/W6781966442","https://openalex.org/W6840337437","https://openalex.org/W6843025442","https://openalex.org/W6847673998","https://openalex.org/W6850961897","https://openalex.org/W6851578965","https://openalex.org/W6860435322"],"related_works":["https://openalex.org/W2000785801","https://openalex.org/W986318368","https://openalex.org/W2384410913","https://openalex.org/W2352878646","https://openalex.org/W2004734601","https://openalex.org/W2130149817","https://openalex.org/W2990194547","https://openalex.org/W1480123525","https://openalex.org/W2620865396","https://openalex.org/W2414054180"],"abstract_inverted_index":{"Abstract":[0],"GEMM":[1,60,80,112],"(General":[2],"Matrix":[3],"Multiplication)":[4],"serves":[5],"as":[6,21,93],"a":[7,58],"fundamental":[8],"operator":[9],"for":[10,150],"deep":[11,17,65],"learning":[12,18,66],"computations.":[13],"Especially":[14],"in":[15,31,43,109,126],"attention-based":[16,127],"models,":[19],"such":[20,92],"Bert,":[22],"GPT,":[23],"and":[24,69,78,130],"SAM,":[25],"the":[26,39,44,52,64,94,103,143],"sizes":[27],"of":[28,47,96,106,119,138,145],"matrices":[29],"involved":[30],"GEMMs":[32],"exhibit":[33],"an":[34,71,115,133],"unbalanced":[35,110],"distribution":[36],"due":[37],"to":[38,76],"variable":[40],"input,":[41],"resulting":[42],"low":[45],"utilization":[46],"hardware":[48,88],"resources.":[49],"To":[50],"address":[51],"issue,":[53],"this":[54],"paper":[55],"proposes":[56],"inserting":[57],"novel":[59],"processing":[61],"layer":[62],"into":[63],"inference":[67,135],"stack":[68],"using":[70],"adaptive":[72],"load":[73],"balancing":[74],"method":[75,84,108,122],"partition":[77],"schedule":[79],"computation":[81,151],"tasks.":[82],"The":[83,121],"is":[85],"implemented":[86],"with":[87],"runtime":[89],"resource":[90],"information,":[91],"occupancy":[95],"computing":[97],"units,":[98],"etc.":[99],"Experiment":[100],"results":[101],"show":[102],"remarkable":[104],"performance":[105,117],"our":[107],"input":[111],"scenarios,":[113],"achieving":[114,132],"average":[116,134],"improvement":[118,137],"2.3x.":[120],"also":[123],"performs":[124],"well":[125],"models":[128],"(GPT-2":[129],"SAM),":[131],"speed":[136],"1.1x.":[139],"These":[140],"findings":[141],"highlight":[142],"effectiveness":[144],"resource-aware":[146],"algorithm":[147],"optimization,":[148],"especially":[149],"task":[152],"scheduling.":[153]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":5}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}
