{"id":"https://openalex.org/W7133520538","doi":"https://doi.org/10.1109/hpca68181.2026.11408495","title":"FlashFuser: Expanding the Scale of Kernel Fusion for Compute-Intensive Operators via Inter-Core Connection","display_name":"FlashFuser: Expanding the Scale of Kernel Fusion for Compute-Intensive Operators via Inter-Core Connection","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7133520538","doi":"https://doi.org/10.1109/hpca68181.2026.11408495"},"language":null,"primary_location":{"id":"doi:10.1109/hpca68181.2026.11408495","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408495","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","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":"Ziyu Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ziyu Huang","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078261857","display_name":"Yangjie Zhou","orcid":"https://orcid.org/0000-0002-3652-5437"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yangjie Zhou","raw_affiliation_strings":["National University of Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128087019","display_name":"Zihan Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zihan Liu","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113192926","display_name":"X. Luo","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinhao Luo","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080651207","display_name":"Yijia Diao","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yijia Diao","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128045567","display_name":"Minyi Guo","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minyi Guo","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jidong Zhai","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jidong Zhai","raw_affiliation_strings":["Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086085686","display_name":"Feng Yu","orcid":"https://orcid.org/0000-0002-4745-5219"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Feng","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128101731","display_name":"Chen Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Zhang","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057379447","display_name":"Anbang Wu","orcid":"https://orcid.org/0009-0000-2596-9385"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Anbang Wu","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003939279","display_name":"Jingwen Leng","orcid":"https://orcid.org/0000-0002-5660-5493"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingwen Leng","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":11,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.50153623,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"14"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.1062999963760376,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.1062999963760376,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.09099999815225601,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.08529999852180481,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/connection","display_name":"Connection (principal bundle)","score":0.5339999794960022},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4943000078201294},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.45579999685287476},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.33739998936653137},{"id":"https://openalex.org/keywords/operator","display_name":"Operator (biology)","score":0.3156000077724457}],"concepts":[{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.5339999794960022},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.49810001254081726},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4943999946117401},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4943000078201294},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.45579999685287476},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.45210000872612},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35339999198913574},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.33739998936653137},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.3156000077724457},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3151000142097473},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2757999897003174},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2703999876976013}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca68181.2026.11408495","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408495","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":45,"referenced_works":["https://openalex.org/W2805566098","https://openalex.org/W2914500262","https://openalex.org/W2954698171","https://openalex.org/W2981758446","https://openalex.org/W3012249773","https://openalex.org/W3089469538","https://openalex.org/W3093933627","https://openalex.org/W3123639931","https://openalex.org/W3159607817","https://openalex.org/W4212986322","https://openalex.org/W4251637954","https://openalex.org/W4281932226","https://openalex.org/W4297097464","https://openalex.org/W4308090436","https://openalex.org/W4312639064","https://openalex.org/W4318541535","https://openalex.org/W4318541538","https://openalex.org/W4318541553","https://openalex.org/W4318541578","https://openalex.org/W4360831828","https://openalex.org/W4366341968","https://openalex.org/W4376456262","https://openalex.org/W4386090486","https://openalex.org/W4388581107","https://openalex.org/W4389491877","https://openalex.org/W4392910976","https://openalex.org/W4394871726","https://openalex.org/W4394871790","https://openalex.org/W4394998892","https://openalex.org/W4395020674","https://openalex.org/W4395106477","https://openalex.org/W4395111944","https://openalex.org/W4400409880","https://openalex.org/W4402427803","https://openalex.org/W4404954401","https://openalex.org/W4405756380","https://openalex.org/W4406157123","https://openalex.org/W4408304436","https://openalex.org/W4409133360","https://openalex.org/W4409248487","https://openalex.org/W4409248600","https://openalex.org/W4413014939","https://openalex.org/W4415797413","https://openalex.org/W4415798989","https://openalex.org/W4416203785"],"related_works":[],"abstract_inverted_index":{"The":[0],"scaling":[1],"of":[2,32,232],"computation":[3],"throughput":[4],"continues":[5],"to":[6,24,39,89,106,121,165,205],"outpace":[7],"improvements":[8],"in":[9,246],"memory":[10,83,168,184,223],"bandwidth,":[11],"making":[12],"many":[13],"deep":[14],"learning":[15],"workloads":[16],"memory-bound.":[17],"Kernel":[18],"fusion":[19,30,56,112,119],"is":[20],"a":[21,77,132,153,192,247],"key":[22],"technique":[23],"alleviate":[25],"this":[26,97],"problem,":[27],"but":[28],"the":[29,45,49,55,62,102,122,166,172,208],"strategies":[31,204],"existing":[33],"compilers":[34],"and":[35,80,148,162,176,201,228,240],"frameworks":[36],"are":[37],"limited":[38,50],"using":[40],"local":[41],"scratchpad":[42],"memory.":[43],"When":[44],"intermediate":[46],"results":[47],"exceed":[48],"capacity":[51],"(such":[52],"as":[53,72,145],"FFN),":[54],"fails.":[57],"Although":[58],"modern":[59,114],"GPUs":[60],"(like":[61],"NVIDIA":[63,216],"H100)":[64],"now":[65],"incorporate":[66],"an":[67,215],"inter-core":[68,108],"connection":[69,109],"mechanism":[70],"known":[71],"Distributed":[73],"Shared":[74],"Memory":[75],"(DSM)\u2014providing":[76],"larger,":[78],"high-bandwidth,":[79],"low-latency":[81],"on-chip":[82],"pool\u2014this":[84],"hardware":[85],"potential":[86],"has":[87],"yet":[88],"be":[90],"exploited":[91],"by":[92,179,225],"software":[93],"frameworks.":[94],"To":[95],"bridge":[96],"gap,":[98],"we":[99,130,151],"present":[100],"FlashFuser,":[101],"first":[103],"compiler":[104],"framework":[105],"utilize":[107],"for":[110],"kernel":[111,230],"on":[113,214],"GPUs.":[115],"FlashFuser":[116,187,221],"extends":[117],"established":[118],"techniques":[120],"DSM":[123],"domain":[124],"through":[125],"three":[126],"core":[127],"contributions.":[128],"First,":[129],"propose":[131],"powerful":[133],"DSM-based":[134],"communication":[135],"abstraction":[136],"that":[137,156,196,220],"formalizes":[138],"complex":[139],"cluster-based":[140],"data":[141,181],"exchange":[142],"patterns,":[143],"such":[144],"reduce,":[146],"shuffle":[147],"multiply.":[149],"Second,":[150],"introduce":[152],"dataflow":[154],"analyzer":[155],"generalizes":[157],"loop":[158],"scheduling,":[159],"resource":[160],"mapping,":[161],"tile":[163,177],"selection":[164],"distributed":[167],"hierarchy;":[169],"it":[170],"determines":[171],"optimal":[173,209],"execution":[174,210],"order":[175],"sizes":[178],"quantifying":[180],"movement":[182],"across":[183],"levels.":[185],"Finally,":[186],"integrates":[188],"these":[189],"components":[190],"into":[191],"unified":[193],"search":[194],"engine":[195],"employs":[197],"analytical":[198],"cost":[199],"modeling":[200],"DSM-aware":[202],"pruning":[203],"efficiently":[206],"discover":[207],"plan.":[211],"Our":[212],"evaluation":[213],"H100":[217],"GPU":[218],"shows":[219],"reduces":[222],"access":[224],"58":[226],"%":[227],"delivers":[229],"speedups":[231],"<tex":[233,248],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[234,249],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$3.3":[235],"x$</tex>":[236],"against":[237,242],"highly-tuned":[238],"libraries":[239],"4.1x":[241],"state-of-the-art":[243],"compilers,":[244],"resulting":[245],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$1.24":[250],"\\times$</tex>":[251],"end-to-end":[252],"speedup.":[253]},"counts_by_year":[],"updated_date":"2026-03-25T23:56:10.502304","created_date":"2026-03-05T00:00:00"}