{"id":"https://openalex.org/W4310362310","doi":"https://doi.org/10.1145/3572917","title":"Scale-out Systolic Arrays","display_name":"Scale-out Systolic Arrays","publication_year":2022,"publication_date":"2022-11-29","ids":{"openalex":"https://openalex.org/W4310362310","doi":"https://doi.org/10.1145/3572917"},"language":"en","primary_location":{"id":"doi:10.1145/3572917","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3572917","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3572917","source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3572917","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047647989","display_name":"Ahmet Caner Y\u00fcz\u00fcg\u00fcler","orcid":"https://orcid.org/0000-0001-7809-9897"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Ahmet Caner Y\u00fcz\u00fcg\u00fcler","raw_affiliation_strings":["EPFL, Lausanne, VD, Switzerland"],"affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, VD, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089938808","display_name":"Canberk S\u00f6nmez","orcid":"https://orcid.org/0000-0003-4542-2947"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Canberk S\u00f6nmez","raw_affiliation_strings":["EPFL, Lausanne, VD, Switzerland"],"affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, VD, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089926912","display_name":"Mario Drumond","orcid":"https://orcid.org/0000-0002-1981-3525"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mario Drumond","raw_affiliation_strings":["CodeDepot, Switzerland"],"affiliations":[{"raw_affiliation_string":"CodeDepot, Switzerland","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015526090","display_name":"Yunho Oh","orcid":"https://orcid.org/0000-0001-6442-3705"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yunho Oh","raw_affiliation_strings":["Korea University, South Korea"],"affiliations":[{"raw_affiliation_string":"Korea University, South Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057697787","display_name":"Babak Falsafi","orcid":"https://orcid.org/0000-0001-5916-8068"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Babak Falsafi","raw_affiliation_strings":["EPFL, Lausanne, VD, Switzerland"],"affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, VD, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5000947076","display_name":"Pascal Frossard","orcid":"https://orcid.org/0000-0002-4010-714X"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Pascal Frossard","raw_affiliation_strings":["EPFL, Lausanne, VD, Switzerland"],"affiliations":[{"raw_affiliation_string":"EPFL, Lausanne, VD, Switzerland","institution_ids":["https://openalex.org/I5124864"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5047647989"],"corresponding_institution_ids":["https://openalex.org/I5124864"],"apc_list":null,"apc_paid":null,"fwci":2.2091,"has_fulltext":true,"cited_by_count":25,"citation_normalized_percentile":{"value":0.88003932,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":"20","issue":"2","first_page":"1","last_page":"25"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10829","display_name":"Interconnection Networks and Systems","score":0.9955999851226807,"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/T10207","display_name":"Advanced biosensing and bioanalysis techniques","score":0.9896000027656555,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/systolic-array","display_name":"Systolic array","score":0.8241898417472839},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.745815634727478},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.6597728133201599},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6231948137283325},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5736044645309448},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.51948082447052},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.44234147667884827},{"id":"https://openalex.org/keywords/interconnection","display_name":"Interconnection","score":0.4258171617984772},{"id":"https://openalex.org/keywords/partition","display_name":"Partition (number theory)","score":0.42290911078453064},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3671765625476837},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.17667603492736816},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.17002403736114502},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.13143819570541382},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.0953533947467804},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09474903345108032},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.0808451771736145}],"concepts":[{"id":"https://openalex.org/C150741067","wikidata":"https://www.wikidata.org/wiki/Q2377218","display_name":"Systolic array","level":3,"score":0.8241898417472839},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.745815634727478},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.6597728133201599},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6231948137283325},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5736044645309448},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.51948082447052},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.44234147667884827},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.4258171617984772},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.42290911078453064},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3671765625476837},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.17667603492736816},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.17002403736114502},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.13143819570541382},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0953533947467804},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09474903345108032},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0808451771736145},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3572917","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3572917","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3572917","source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},{"id":"pmh:oai:infoscience.epfl.ch:303292","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/198683","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"WoS","raw_type":"research article"}],"best_oa_location":{"id":"doi:10.1145/3572917","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3572917","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3572917","source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4310362310.pdf","grobid_xml":"https://content.openalex.org/works/W4310362310.grobid-xml"},"referenced_works_count":53,"referenced_works":["https://openalex.org/W2017369466","https://openalex.org/W2114870379","https://openalex.org/W2117696986","https://openalex.org/W2125203716","https://openalex.org/W2152839228","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2233797083","https://openalex.org/W2289252105","https://openalex.org/W2498311662","https://openalex.org/W2503158931","https://openalex.org/W2604319603","https://openalex.org/W2605347906","https://openalex.org/W2606722458","https://openalex.org/W2612076670","https://openalex.org/W2790925711","https://openalex.org/W2794670651","https://openalex.org/W2883929540","https://openalex.org/W2900228909","https://openalex.org/W2900327659","https://openalex.org/W2906043559","https://openalex.org/W2935331687","https://openalex.org/W2943476754","https://openalex.org/W2945146780","https://openalex.org/W2945580137","https://openalex.org/W2949660525","https://openalex.org/W2950656546","https://openalex.org/W2962987932","https://openalex.org/W2963341956","https://openalex.org/W2963446712","https://openalex.org/W2965261596","https://openalex.org/W2972054167","https://openalex.org/W2980020162","https://openalex.org/W2980104813","https://openalex.org/W2980200167","https://openalex.org/W2982960593","https://openalex.org/W3012178976","https://openalex.org/W3016542674","https://openalex.org/W3016939927","https://openalex.org/W3036878841","https://openalex.org/W3043406639","https://openalex.org/W3083448369","https://openalex.org/W3118417089","https://openalex.org/W3130554079","https://openalex.org/W3148444620","https://openalex.org/W3190062760","https://openalex.org/W3206621799","https://openalex.org/W4245199738","https://openalex.org/W4247353671","https://openalex.org/W4280635517","https://openalex.org/W4282008392","https://openalex.org/W4285257701","https://openalex.org/W6756118020"],"related_works":["https://openalex.org/W2931688134","https://openalex.org/W2377919138","https://openalex.org/W2378857091","https://openalex.org/W4256502920","https://openalex.org/W103652678","https://openalex.org/W2999756192","https://openalex.org/W4226090359","https://openalex.org/W2383090288","https://openalex.org/W2387846155","https://openalex.org/W1822341218"],"abstract_inverted_index":{"Multi-pod":[0],"systolic":[1,19,47],"arrays":[2,20],"are":[3],"emerging":[4],"as":[5],"the":[6,79],"architecture":[7],"of":[8,35,99,146,167],"choice":[9],"in":[10,45,82,116,151],"DNN":[11,156],"inference":[12,128,157],"accelerators.":[13],"Despite":[14],"their":[15],"potential,":[16],"designing":[17],"multi-pod":[18,46,127,162],"to":[21,113,148],"maximize":[22,114],"effective":[23,152],"throughput/Watt\u2014i.e.,":[24],"throughput/Watt":[25],"adjusted":[26],"when":[27],"accounting":[28],"for":[29,73,93,130,154],"array":[30,48,51,59,71],"utilization\u2014poses":[31],"a":[32,90,96,104,126,165],"unique":[33],"set":[34],"challenges.":[36],"In":[37],"this":[38],"work,":[39],"we":[40,102],"study":[41],"three":[42,138],"key":[43],"pillars":[44],"designs,":[49],"namely":[50],"granularity,":[52],"interconnect,":[53],"and":[54,63,84,133,159],"tiling.":[55],"We":[56,120,140],"identify":[57],"optimal":[58],"granularity":[60],"across":[61],"workloads":[62],"show":[64,85,141],"that":[65,86,142],"state-of-the-art":[66,155,161],"commercial":[67],"accelerators":[68,94,163],"use":[69],"suboptimal":[70],"sizes":[72],"single-tenancy":[74],"workloads.":[75],"We,":[76],"then":[77],"evaluate":[78],"bandwidth/latency":[80],"trade-offs":[81],"interconnects":[83],"Butterfly":[87],"networks":[88],"offer":[89],"scalable":[91],"topology":[92],"with":[95,109],"large":[97],"number":[98],"pods.":[100,119],"Finally,":[101],"introduce":[103],"novel":[105],"data":[106],"tiling":[107],"scheme":[108],"custom":[110],"partition":[111],"size":[112],"utilization":[115],"optimally":[117],"sized":[118],"propose":[121],"Scale-out":[122],"Systolic":[123],"Arrays":[124],",":[125],"accelerator":[129],"both":[131],"single-":[132],"multi-tenancy":[134],"based":[135],"on":[136],"these":[137],"pillars.":[139],"SOSA":[143],"exhibits":[144],"scaling":[145],"up":[147],"600":[149],"TeraOps/s":[150],"throughput":[153],"workloads,":[158],"outperforms":[160],"by":[164],"factor":[166],"1.5":[168],"\u00d7.":[169],"1":[170]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":8},{"year":2023,"cited_by_count":4}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}