{"id":"https://openalex.org/W3089019106","doi":"https://doi.org/10.1145/3499757","title":"E-BATCH: Energy-Efficient and High-Throughput RNN Batching","display_name":"E-BATCH: Energy-Efficient and High-Throughput RNN Batching","publication_year":2022,"publication_date":"2022-01-23","ids":{"openalex":"https://openalex.org/W3089019106","doi":"https://doi.org/10.1145/3499757","mag":"3089019106"},"language":"en","primary_location":{"id":"doi:10.1145/3499757","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3499757","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3499757","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":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3499757","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5078459594","display_name":"Franyell Silfa","orcid":"https://orcid.org/0000-0002-1134-9908"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Franyell Silfa","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032189355","display_name":"Jos\u00e9-Mar\u00eda Arnau","orcid":null},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Jose Maria Arnau","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100733331","display_name":"Antonio Gonz\u00e1lez","orcid":"https://orcid.org/0000-0002-0009-0996"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Antonio Gonz\u00e1lez","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5078459594"],"corresponding_institution_ids":["https://openalex.org/I9617848"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.00240142,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"19","issue":"1","first_page":"1","last_page":"23"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9965000152587891,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9959999918937683,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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":0.8285254240036011},{"id":"https://openalex.org/keywords/padding","display_name":"Padding","score":0.7736684083938599},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.7315948605537415},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6749317646026611},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.6119926571846008},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.5929639935493469},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.465131938457489},{"id":"https://openalex.org/keywords/batch-processing","display_name":"Batch processing","score":0.4414169192314148},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.42077553272247314},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.4032072126865387},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3643527328968048},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.29093560576438904},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.18098244071006775},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.10542884469032288},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.06253054738044739}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8285254240036011},{"id":"https://openalex.org/C165435473","wikidata":"https://www.wikidata.org/wiki/Q1509884","display_name":"Padding","level":2,"score":0.7736684083938599},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.7315948605537415},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6749317646026611},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.6119926571846008},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.5929639935493469},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.465131938457489},{"id":"https://openalex.org/C172658912","wikidata":"https://www.wikidata.org/wiki/Q661613","display_name":"Batch processing","level":2,"score":0.4414169192314148},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.42077553272247314},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.4032072126865387},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3643527328968048},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.29093560576438904},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.18098244071006775},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.10542884469032288},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.06253054738044739},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3499757","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3499757","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3499757","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:arXiv.org:2009.10656","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2009.10656","pdf_url":"https://arxiv.org/pdf/2009.10656","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:3089019106","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2009.10656","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2009.10656","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2009.10656","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.1145/3499757","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3499757","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3499757","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":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.9100000262260437}],"awards":[{"id":"https://openalex.org/G3025668325","display_name":null,"funder_award_id":"RB-I00","funder_id":"https://openalex.org/F4320335598","funder_display_name":"Agencia Estatal de Investigaci\u00f3n"},{"id":"https://openalex.org/G3429648993","display_name":null,"funder_award_id":"PID202","funder_id":"https://openalex.org/F4320335598","funder_display_name":"Agencia Estatal de Investigaci\u00f3n"},{"id":"https://openalex.org/G7667637836","display_name":null,"funder_award_id":"PID2020-113172RB-I00","funder_id":"https://openalex.org/F4320335598","funder_display_name":"Agencia Estatal de Investigaci\u00f3n"}],"funders":[{"id":"https://openalex.org/F4320322121","display_name":"Instituci\u00f3 Catalana de Recerca i Estudis Avan\u00e7ats","ror":"https://ror.org/0371hy230"},{"id":"https://openalex.org/F4320335598","display_name":"Agencia Estatal de Investigaci\u00f3n","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3089019106.pdf","grobid_xml":"https://content.openalex.org/works/W3089019106.grobid-xml"},"referenced_works_count":36,"referenced_works":["https://openalex.org/W1667652561","https://openalex.org/W1909318870","https://openalex.org/W2024001079","https://openalex.org/W2064675550","https://openalex.org/W2139542266","https://openalex.org/W2157331557","https://openalex.org/W2186615578","https://openalex.org/W2271840356","https://openalex.org/W2402144811","https://openalex.org/W2463955103","https://openalex.org/W2464177207","https://openalex.org/W2513383847","https://openalex.org/W2525778437","https://openalex.org/W2585720638","https://openalex.org/W2588448445","https://openalex.org/W2594990650","https://openalex.org/W2606722458","https://openalex.org/W2770159745","https://openalex.org/W2798291715","https://openalex.org/W2798405286","https://openalex.org/W2883899390","https://openalex.org/W2883929540","https://openalex.org/W2886885214","https://openalex.org/W2899771611","https://openalex.org/W2911491685","https://openalex.org/W2934853022","https://openalex.org/W2949640717","https://openalex.org/W2953212265","https://openalex.org/W2963211739","https://openalex.org/W2963316143","https://openalex.org/W2980084320","https://openalex.org/W2984140583","https://openalex.org/W2986441544","https://openalex.org/W3105037130","https://openalex.org/W3125484574","https://openalex.org/W6719768283"],"related_works":["https://openalex.org/W2107954672","https://openalex.org/W995689611","https://openalex.org/W2466635064","https://openalex.org/W2896090304","https://openalex.org/W2905143468","https://openalex.org/W2553243469","https://openalex.org/W2113243322","https://openalex.org/W2073593819","https://openalex.org/W2185353116","https://openalex.org/W2767520126","https://openalex.org/W2128898110","https://openalex.org/W2275552531","https://openalex.org/W2771229894","https://openalex.org/W2061003956","https://openalex.org/W2563587242","https://openalex.org/W2132839180","https://openalex.org/W430285124","https://openalex.org/W2155783364","https://openalex.org/W2181874803","https://openalex.org/W2102452869"],"abstract_inverted_index":{"Recurrent":[0],"Neural":[1],"Network":[2],"(RNN)":[3],"inference":[4],"exhibits":[5],"low":[6],"hardware":[7,90,162],"utilization":[8],"due":[9],"to":[10,79,97,129,149],"the":[11,33,46,107,112,135,142,151,160,199],"strict":[12],"data":[13],"dependencies":[14],"across":[15],"time-steps.":[16],"Batching":[17],"multiple":[18,95],"requests":[19],"can":[20,125],"increase":[21],"throughput.":[22],"However,":[23,53],"RNN":[24,58,80],"batching":[25,76],"requires":[26],"a":[27,61,72,85,121,130],"large":[28,99],"amount":[29,136],"of":[30,84,114,137,144,169],"padding":[31],"since":[32],"batched":[34],"input":[35,116,123],"sequences":[36,96],"may":[37],"vastly":[38],"differ":[39],"in":[40,60,102,183,186],"length.":[41],"Schemes":[42],"that":[43],"dynamically":[44,140],"update":[45],"batch":[47,148],"every":[48],"few":[49],"time-steps":[50,145],"avoid":[51],"padding.":[52,138],"they":[54],"require":[55],"executing":[56],"different":[57],"layers":[59],"short":[62],"time":[63],"span,":[64],"decreasing":[65],"energy":[66,104,157,179,194],"efficiency.":[67],"Hence,":[68,120],"we":[69],"propose":[70],"E-BATCH,":[71],"low-latency":[73],"and":[74,88,156,171,178,193],"energy-efficient":[75],"scheme":[77],"tailored":[78],"accelerators.":[81],"It":[82],"consists":[83],"runtime":[86,93],"system":[87],"effective":[89],"support.":[91],"The":[92],"concatenates":[94],"create":[98],"batches,":[100],"resulting":[101],"substantial":[103],"savings.":[105],"Furthermore,":[106],"accelerator":[108],"notifies":[109],"it":[110,188],"when":[111],"evaluation":[113],"an":[115],"sequence":[117,124],"is":[118],"done.":[119],"new":[122],"be":[126],"immediately":[127],"added":[128],"batch,":[131],"thus":[132],"largely":[133],"reducing":[134],"E-BATCH":[139,166,173],"controls":[141],"number":[143],"evaluated":[146],"per":[147],"achieve":[150],"best":[152],"trade-off":[153],"between":[154],"latency":[155],"efficiency":[158,180,195],"for":[159],"given":[161],"platform.":[163],"We":[164],"evaluate":[165],"on":[167],"top":[168],"E-PUR":[170],"TPU.":[172],"improves":[174,189],"throughput":[175,190],"by":[176,181,191,196],"1.8\u00d7":[177],"3.6\u00d7":[182],"E-PUR,":[184],"whereas":[185],"TPU,":[187],"2.1\u00d7":[192],"1.6\u00d7,":[197],"over":[198],"state-of-the-art.":[200]},"counts_by_year":[{"year":2020,"cited_by_count":1}],"updated_date":"2026-03-13T16:22:10.518609","created_date":"2025-10-10T00:00:00"}