{"id":"https://openalex.org/W2788418609","doi":"https://doi.org/10.1145/3174243.3174964","title":"An Optimal Microarchitecture for Stencil Computation with Data Reuse and Fine-Grained Parallelism","display_name":"An Optimal Microarchitecture for Stencil Computation with Data Reuse and Fine-Grained Parallelism","publication_year":2018,"publication_date":"2018-02-15","ids":{"openalex":"https://openalex.org/W2788418609","doi":"https://doi.org/10.1145/3174243.3174964","mag":"2788418609"},"language":"en","primary_location":{"id":"doi:10.1145/3174243.3174964","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3174243.3174964","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","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/A5074835029","display_name":"Yuze Chi","orcid":"https://orcid.org/0000-0002-5885-0425"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yuze Chi","raw_affiliation_strings":["University of California, Los Angeles, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"University of California, Los Angeles, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063866156","display_name":"Peipei Zhou","orcid":"https://orcid.org/0000-0002-0493-1844"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peipei Zhou","raw_affiliation_strings":["University of California, Los Angeles, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"University of California, Los Angeles, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102170602","display_name":"Jason Cong","orcid":null},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jason Cong","raw_affiliation_strings":["University of California, Los Angeles, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"University of California, Los Angeles, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5074835029"],"corresponding_institution_ids":["https://openalex.org/I161318765"],"apc_list":null,"apc_paid":null,"fwci":0.2525,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.45710095,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"286","last_page":"286"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9991999864578247,"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":0.9991999864578247,"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":0.9890000224113464,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.9879000186920166,"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/computer-science","display_name":"Computer science","score":0.8618295192718506},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7353255152702332},{"id":"https://openalex.org/keywords/stencil","display_name":"Stencil","score":0.7050543427467346},{"id":"https://openalex.org/keywords/microarchitecture","display_name":"Microarchitecture","score":0.6232078075408936},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.5557404160499573},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4643673300743103},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.43785354495048523},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3558063805103302},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.13701969385147095}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8618295192718506},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7353255152702332},{"id":"https://openalex.org/C76752949","wikidata":"https://www.wikidata.org/wiki/Q7607499","display_name":"Stencil","level":2,"score":0.7050543427467346},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.6232078075408936},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.5557404160499573},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4643673300743103},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.43785354495048523},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3558063805103302},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.13701969385147095},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3174243.3174964","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3174243.3174964","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.8700000047683716}],"awards":[{"id":"https://openalex.org/G3318889966","display_name":null,"funder_award_id":"20134321","funder_id":"https://openalex.org/F4320307102","funder_display_name":"Intel Corporation"},{"id":"https://openalex.org/G4595116112","display_name":null,"funder_award_id":"3002565073","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"}],"funders":[{"id":"https://openalex.org/F4320307102","display_name":"Intel Corporation","ror":"https://ror.org/01ek73717"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2076915000","https://openalex.org/W2769005600","https://openalex.org/W2620649220","https://openalex.org/W4243297508","https://openalex.org/W4376647684","https://openalex.org/W2097757554","https://openalex.org/W2039378765","https://openalex.org/W2185562986","https://openalex.org/W4401408752","https://openalex.org/W2056599241"],"abstract_inverted_index":{"Stencil":[0],"computation":[1],"is":[2,30],"one":[3],"of":[4,37,88,95,108,132,174,225],"the":[5,73,80,86,93,103,106,144],"most":[6],"important":[7],"kernels":[8],"for":[9,61,199],"many":[10],"applications":[11,200],"such":[12],"as":[13],"image":[14],"processing,":[15],"solving":[16],"partial":[17],"differential":[18],"equations,":[19],"and":[20,42,53,56,77,98,158,214],"cellular":[21],"automata.":[22],"Nevertheless,":[23],"implementing":[24],"a":[25,125,149,195,226],"high":[26,38,211],"throughput":[27,224],"stencil":[28,62],"kernel":[29],"not":[31,69],"trivial":[32],"due":[33,208],"to":[34,113,154,209],"its":[35,210],"nature":[36],"memory":[39,75,118,212],"access":[40],"load":[41],"low":[43,215],"operational":[44],"intensity.":[45],"In":[46],"this":[47,122],"work":[48],"we":[49,147],"adopt":[50],"data":[51,65,89],"reuse":[52,66,79,90],"fine-grained":[54,96],"parallelism":[55],"present":[57],"an":[58,172],"optimal":[59],"microarchitecture":[60,123],"computation.":[63],"The":[64],"line":[67,178],"buffers":[68],"only":[70,182],"fully":[71,78,99,196],"utilize":[72],"external":[74],"bandwidth":[76],"input":[81],"data,":[82],"they":[83],"also":[84],"minimize":[85],"size":[87],"buffer":[91],"given":[92],"number":[94,107],"parallelized":[97],"pipelined":[100,197],"PEs.":[101],"With":[102],"proposed":[104],"microarchitecture,":[105],"PEs":[109],"can":[110,170,193],"be":[111,203],"increased":[112],"saturate":[114],"all":[115],"available":[116],"off-chip":[117],"bandwidth.":[119],"We":[120],"implement":[121,194],"with":[124,181,187,205,229],"high-level":[126],"synthesis":[127],"(HLS)":[128],"based":[129],"template":[130],"instead":[131],"register":[133],"transfer":[134],"level":[135],"(RTL)":[136],"specifications,":[137],"which":[138],"provides":[139,222],"great":[140],"programmability.":[141],"To":[142],"guide":[143],"system":[145],"design,":[146,190],"propose":[148],"performance":[150],"model":[151,156],"in":[152],"addition":[153],"detailed":[155],"evaluation":[157],"optimization":[159],"analysis.":[160],"Experimental":[161],"results":[162],"from":[163],"on-board":[164],"execution":[165],"show":[166],"that":[167,201],"our":[168,191,219],"design":[169,180,192,216],"provide":[171],"average":[173],"6.5x":[175],"speedup":[176],"over":[177],"buffer-only":[179],"2.4x":[183],"resource":[184],"overhead.":[185],"Compared":[186],"loop":[188,206],"transformation-only":[189,207],"accelerator":[198],"cannot":[202],"implemented":[204],"conflict":[213],"flexibility.":[217],"Furthermore,":[218],"FPGA":[220],"implementation":[221],"83%":[223],"14-core":[227],"CPU":[228],"4x":[230],"energy-efficiency.":[231]},"counts_by_year":[{"year":2018,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}