{"id":"https://openalex.org/W3091523831","doi":"https://doi.org/10.1145/3410463.3414628","title":"Opportunistic Early Pipeline Re-steering for Data-dependent Branches","display_name":"Opportunistic Early Pipeline Re-steering for Data-dependent Branches","publication_year":2020,"publication_date":"2020-09-30","ids":{"openalex":"https://openalex.org/W3091523831","doi":"https://doi.org/10.1145/3410463.3414628","mag":"3091523831"},"language":"en","primary_location":{"id":"doi:10.1145/3410463.3414628","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3410463.3414628","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM International Conference on Parallel Architectures and Compilation Techniques","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/A5064269047","display_name":"Saurabh Gupta","orcid":"https://orcid.org/0000-0002-5984-852X"},"institutions":[{"id":"https://openalex.org/I4210146682","display_name":"Intel (India)","ror":"https://ror.org/04f2n1245","country_code":"IN","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210146682"]}],"countries":["IN"],"is_corresponding":true,"raw_author_name":"Saurabh Gupta","raw_affiliation_strings":["Intel Labs, Intel Corp., Bengaluru, India"],"affiliations":[{"raw_affiliation_string":"Intel Labs, Intel Corp., Bengaluru, India","institution_ids":["https://openalex.org/I4210146682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027273390","display_name":"Niranjan Soundararajan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210146682","display_name":"Intel (India)","ror":"https://ror.org/04f2n1245","country_code":"IN","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210146682"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Niranjan Soundararajan","raw_affiliation_strings":["Intel Labs, Intel Corp., Bengaluru, India"],"affiliations":[{"raw_affiliation_string":"Intel Labs, Intel Corp., Bengaluru, India","institution_ids":["https://openalex.org/I4210146682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103563776","display_name":"Ragavendra Natarajan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210146682","display_name":"Intel (India)","ror":"https://ror.org/04f2n1245","country_code":"IN","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210146682"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ragavendra Natarajan","raw_affiliation_strings":["Intel Corp., Bengaluru, India"],"affiliations":[{"raw_affiliation_string":"Intel Corp., Bengaluru, India","institution_ids":["https://openalex.org/I4210146682"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036799188","display_name":"Sreenivas Subramoney","orcid":"https://orcid.org/0000-0001-5372-0173"},"institutions":[{"id":"https://openalex.org/I4210146682","display_name":"Intel (India)","ror":"https://ror.org/04f2n1245","country_code":"IN","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210146682"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sreenivas Subramoney","raw_affiliation_strings":["Intel Labs, Intel Corp., Bengaluru, India"],"affiliations":[{"raw_affiliation_string":"Intel Labs, Intel Corp., Bengaluru, India","institution_ids":["https://openalex.org/I4210146682"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5064269047"],"corresponding_institution_ids":["https://openalex.org/I4210146682"],"apc_list":null,"apc_paid":null,"fwci":0.9241,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.72996516,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"305","last_page":"316"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":1.0,"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":1.0,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9977999925613403,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9961000084877014,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.857651948928833},{"id":"https://openalex.org/keywords/speculative-execution","display_name":"Speculative execution","score":0.6774644255638123},{"id":"https://openalex.org/keywords/branch-predictor","display_name":"Branch predictor","score":0.6748765110969543},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.6690293550491333},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6466109752655029},{"id":"https://openalex.org/keywords/microarchitecture","display_name":"Microarchitecture","score":0.4567807912826538},{"id":"https://openalex.org/keywords/out-of-order-execution","display_name":"Out-of-order execution","score":0.44294473528862},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.20953109860420227}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.857651948928833},{"id":"https://openalex.org/C141331961","wikidata":"https://www.wikidata.org/wiki/Q2164465","display_name":"Speculative execution","level":2,"score":0.6774644255638123},{"id":"https://openalex.org/C168522837","wikidata":"https://www.wikidata.org/wiki/Q679552","display_name":"Branch predictor","level":2,"score":0.6748765110969543},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6690293550491333},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6466109752655029},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.4567807912826538},{"id":"https://openalex.org/C1793878","wikidata":"https://www.wikidata.org/wiki/Q1153762","display_name":"Out-of-order execution","level":2,"score":0.44294473528862},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.20953109860420227}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3410463.3414628","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3410463.3414628","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM International Conference on Parallel Architectures and Compilation Techniques","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.9100000262260437,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W198451133","https://openalex.org/W361445374","https://openalex.org/W1978301594","https://openalex.org/W2011784170","https://openalex.org/W2013083013","https://openalex.org/W2035472111","https://openalex.org/W2052312648","https://openalex.org/W2095977058","https://openalex.org/W2097586580","https://openalex.org/W2111740117","https://openalex.org/W2118811116","https://openalex.org/W2148635054","https://openalex.org/W2156996403","https://openalex.org/W2164148722","https://openalex.org/W2244769533","https://openalex.org/W2601699999","https://openalex.org/W2794591174","https://openalex.org/W2883051445","https://openalex.org/W2904243206","https://openalex.org/W2904519778","https://openalex.org/W2913330047","https://openalex.org/W2963163273","https://openalex.org/W4300932323","https://openalex.org/W6843447886"],"related_works":["https://openalex.org/W2139338109","https://openalex.org/W2897302968","https://openalex.org/W2745771918","https://openalex.org/W1916582918","https://openalex.org/W3194312717","https://openalex.org/W2434762079","https://openalex.org/W1595275176","https://openalex.org/W2043979743","https://openalex.org/W37057355","https://openalex.org/W1584635396"],"abstract_inverted_index":{"As":[0,249],"Out-of-Order":[1],"(OOO)":[2],"cores":[3,251],"scale":[4,113],"to":[5,10,36,44,63,74,97,106,127,146,181,204],"very":[6],"large":[7],"instruction":[8,13,139],"windows":[9],"extract":[11],"higher":[12],"level":[14],"parallelism":[15],"(ILP),":[16],"the":[17,27,30,37,46,50,54,65,71,76,80,98,104,148,152,162,199,202,206,218,259,268,271],"cost":[18,78],"of":[19,29,56,79,161,178,201,270],"mis-speculation":[20],"increases":[21],"tremendously.":[22],"Branch":[23],"predictors":[24],"sitting":[25],"at":[26,111],"head":[28],"pipeline":[31,203],"are":[32,89,155],"a":[33,112,124,158,176,211,228],"significant":[34,159],"contributor":[35],"speculatively":[38],"executed":[39],"code.":[40],"It":[41],"is":[42,103,275],"critical":[43],"limit":[45],"execution":[47],"time":[48],"down":[49],"wrong":[51],"path":[52],"for":[53,157],"sake":[55],"performance":[57],"and":[58,69,93,122,142,254],"energy":[59],"efficiency.":[60],"Architects":[61],"continue":[62],"increase":[64],"branch":[66,149,193,207,220,232,273],"predictor":[67,274],"sizes":[68],"improve":[70],"prediction":[72],"algorithms":[73],"mitigate":[75],"increasing":[77],"mis-speculations.":[81],"Unfortunately,":[82],"there":[83],"still":[84],"exists":[85],"branches":[86,110,133,168],"whose":[87],"outcomes":[88],"predominantly":[90],"data":[91,108],"dependent,":[92],"that":[94,132,196,224,258],"significantly":[95],"contribute":[96],"overall":[99,163],"mispredictions.":[100,164],"This":[101],"work":[102],"first":[105],"characterize":[107],"dependent":[109],"spanning":[114],"100+":[115],"workloads":[116],"drawn":[117],"from":[118,151,184,261],"several":[119],"application":[120],"categories":[121],"establish":[123],"clear":[125],"motivation":[126],"address":[128],"them.":[129],"We":[130,165,174,187],"find":[131,223],"which":[134],"have":[135],"only":[136,143],"one":[137],"load":[138,153],"feeding":[140],"them":[141],"simple":[144],"operations":[145],"compute":[147],"direction":[150],"value":[154],"responsible":[156],"fraction":[160],"call":[166],"such":[167,185],"Direct":[169],"Data":[170],"Dependent":[171],"(3D)":[172],"branches.":[173],"develop":[175],"family":[177],"synergistic":[179],"techniques":[180],"avoid":[182],"mispredictions":[183,233],"3D-branches.":[186],"describe":[188],"3D-Branch":[189,262],"Overrider,":[190],"our":[191,225],"novel":[192],"overriding":[194],"technique":[195,226],"operates":[197],"in":[198,231,235,246],"front-end":[200],"minimize":[205],"misprediction":[208],"penalties.":[209],"On":[210],"modern":[212],"Icelake-like":[213],"OOO":[214],"core":[215],"equipped":[216],"with":[217],"state-of-the-art":[219],"predictor,":[221],"we":[222,256],"provides":[227],"12.7%":[229],"reduction":[230],"resulting":[234],"3.1%":[236],"Instructions":[237],"Per":[238],"Cycle":[239],"(IPC)":[240],"gain":[241],"while":[242],"needing":[243],"just":[244],"5.7KB":[245],"additional":[247],"storage.":[248],"future":[250],"become":[252],"wider":[253],"deeper,":[255],"show":[257],"gains":[260],"Overrider":[263],"nicely":[264],"increases,":[265],"even":[266],"if":[267],"size":[269],"underlying":[272],"aggressively":[276],"scaled.":[277]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
