{"id":"https://openalex.org/W2917307783","doi":"https://doi.org/10.1145/3289602.3293947","title":"Overcoming Data Transfer Bottlenecks in DNN Accelerators via Layer-Conscious Memory Managment","display_name":"Overcoming Data Transfer Bottlenecks in DNN Accelerators via Layer-Conscious Memory Managment","publication_year":2019,"publication_date":"2019-02-20","ids":{"openalex":"https://openalex.org/W2917307783","doi":"https://doi.org/10.1145/3289602.3293947","mag":"2917307783"},"language":"en","primary_location":{"id":"doi:10.1145/3289602.3293947","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3289602.3293947","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","raw_type":"proceedings-article"},"type":"conference-abstract","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/A5102904614","display_name":"Xuechao Wei","orcid":"https://orcid.org/0000-0002-0996-2260"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuechao Wei","raw_affiliation_strings":["Peking University & Falcon Computing Solutions, Inc., Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University & Falcon Computing Solutions, Inc., Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100604860","display_name":"Yun Liang","orcid":"https://orcid.org/0000-0002-9076-7998"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yun Liang","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100364062","display_name":"Peng Zhang","orcid":"https://orcid.org/0000-0001-9428-9340"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Zhang","raw_affiliation_strings":["Falcon Computing Solutions, Inc., Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions, Inc., Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008633310","display_name":"Cody Hao Yu","orcid":"https://orcid.org/0000-0002-9298-6254"},"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":"Cody Hao Yu","raw_affiliation_strings":["University of California, Los Angeles, Los Angeles, CA, USA"],"raw_orcid":null,"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/A5043222165","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":["Peking University, University of California, Los Angeles, & Falcon Computing Solutions, Inc., Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University, University of California, Los Angeles, & Falcon Computing Solutions, Inc., Los Angeles, CA, USA","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"120","last_page":"120"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9997000098228455,"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":0.9997000098228455,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.996999979019165,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9952999949455261,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7852328419685364},{"id":"https://openalex.org/keywords/registered-memory","display_name":"Registered memory","score":0.6403141021728516},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4803948998451233},{"id":"https://openalex.org/keywords/interleaved-memory","display_name":"Interleaved memory","score":0.47539982199668884},{"id":"https://openalex.org/keywords/chip","display_name":"Chip","score":0.46664953231811523},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.45083463191986084},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.43147939443588257},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.4298820495605469},{"id":"https://openalex.org/keywords/auxiliary-memory","display_name":"Auxiliary memory","score":0.42681053280830383},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.42501410841941833},{"id":"https://openalex.org/keywords/memory-map","display_name":"Memory map","score":0.42097708582878113},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.41563546657562256},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.37362581491470337},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3672228753566742},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3110012412071228},{"id":"https://openalex.org/keywords/shared-memory","display_name":"Shared memory","score":0.27480366826057434},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.26025304198265076},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.12493249773979187},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.09325021505355835}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7852328419685364},{"id":"https://openalex.org/C93446704","wikidata":"https://www.wikidata.org/wiki/Q449328","display_name":"Registered memory","level":3,"score":0.6403141021728516},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4803948998451233},{"id":"https://openalex.org/C63511323","wikidata":"https://www.wikidata.org/wiki/Q908936","display_name":"Interleaved memory","level":4,"score":0.47539982199668884},{"id":"https://openalex.org/C165005293","wikidata":"https://www.wikidata.org/wiki/Q1074500","display_name":"Chip","level":2,"score":0.46664953231811523},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.45083463191986084},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.43147939443588257},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.4298820495605469},{"id":"https://openalex.org/C82687282","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Auxiliary memory","level":2,"score":0.42681053280830383},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.42501410841941833},{"id":"https://openalex.org/C74426580","wikidata":"https://www.wikidata.org/wiki/Q719484","display_name":"Memory map","level":3,"score":0.42097708582878113},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.41563546657562256},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.37362581491470337},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3672228753566742},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3110012412071228},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.27480366826057434},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.26025304198265076},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.12493249773979187},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.09325021505355835},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3289602.3293947","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3289602.3293947","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 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/9","score":0.4399999976158142,"display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W3021597805","https://openalex.org/W2564569739","https://openalex.org/W1975444747","https://openalex.org/W2062140197","https://openalex.org/W4232365528","https://openalex.org/W2044064773","https://openalex.org/W1975901677","https://openalex.org/W5673888","https://openalex.org/W1975639590","https://openalex.org/W2133063415"],"abstract_inverted_index":{"Deep":[0],"Neural":[1],"Networks":[2],"(DNNs)":[3],"are":[4],"rapidly":[5],"evolving":[6],"to":[7,128,190],"satisfy":[8],"the":[9,53,66,74,87,120,125,140,146,159,172,195],"performance":[10,75],"and":[11,24,34,59,135,199],"accuracy":[12],"requirements":[13,130],"in":[14,27,93,97],"many":[15],"real":[16],"world":[17],"applications.":[18],"The":[19],"evolution":[20],"renders":[21],"DNNs":[22],"more":[23,25],"complex":[26],"terms":[28],"of":[29,56,89,99,123,150,161],"network":[30],"topology,":[31],"data":[32,54,90,141,160],"sizes":[33],"layer":[35],"types.":[36],"Currently":[37],"most":[38],"state-of-the-art":[39,203],"DNN":[40,115],"accelerators":[41,173],"adopt":[42],"a":[43,108,144],"uniform":[44],"memory":[45,68,81,110,121,133,148,151,185],"hierarchy":[46,111],"(UMH)":[47],"design":[48],"methodology,":[49],"which":[50],"means":[51],"that":[52,180],"transferring":[55],"all":[57,124],"convolutional":[58],"fully":[60],"connected":[61],"layers":[62,126,153],"must":[63],"go":[64],"through":[65],"same":[67],"levels.":[69],"Unfortunately,":[70],"for":[71,114,131,139,171],"some":[72],"layers,":[73],"is":[76,84],"always":[77],"bounded":[78,152],"by":[79,86,157],"off-chip":[80,132,147],"transferring.":[82],"It":[83],"caused":[85],"saturating":[88],"reuse":[91],"happening":[92],"on-chip":[94,100,136],"buffers,":[95],"resulting":[96],"underutilization":[98],"memory.":[101],"To":[102],"address":[103],"this":[104],"issue,":[105],"we":[106,167],"propose":[107],"layer-conscious":[109],"(LCMH)":[112],"methodology":[113],"accelerators.":[116],"LCMH":[117],"could":[118,154,187],"determine":[119],"levels":[122],"according":[127],"their":[129],"bandwidth":[134],"buffer":[137],"size":[138],"sources.":[142],"As":[143],"result,":[145],"footprints":[149],"be":[155],"avoided":[156],"keeping":[158],"them":[162],"on":[163],"chip.":[164],"In":[165],"addition,":[166],"provide":[168],"architectural":[169],"support":[170],"equipped":[174],"with":[175,182,194],"LCMH.":[176],"Experimental":[177],"results":[178],"show":[179],"designs":[181,196],"layer-":[183],"conscious":[184],"management":[186],"achieve":[188],"up":[189],"36%":[191],"speedup":[192],"compared":[193],"wth":[197],"UMH":[198],"5%":[200],"improvement":[201],"over":[202],"designs.":[204]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":3},{"year":2019,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}