{"id":"https://openalex.org/W7125828737","doi":"https://doi.org/10.1109/jetcas.2026.3657823","title":"A Coalesced Tensor Reduction Architecture for Scalable All-Bank PIM Execution","display_name":"A Coalesced Tensor Reduction Architecture for Scalable All-Bank PIM Execution","publication_year":2026,"publication_date":"2026-01-27","ids":{"openalex":"https://openalex.org/W7125828737","doi":"https://doi.org/10.1109/jetcas.2026.3657823"},"language":null,"primary_location":{"id":"doi:10.1109/jetcas.2026.3657823","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jetcas.2026.3657823","pdf_url":null,"source":{"id":"https://openalex.org/S142323794","display_name":"IEEE Journal on Emerging and Selected Topics in Circuits and Systems","issn_l":"2156-3357","issn":["2156-3357","2156-3365"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal on Emerging and Selected Topics in Circuits and Systems","raw_type":"journal-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/A5080915893","display_name":"Taehyung Park","orcid":"https://orcid.org/0000-0003-0631-3027"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Taehyung Park","raw_affiliation_strings":["Department of Electrical Engineering and Computer Science, Seoul National University, Inter-University Semiconductor Research Center (ISRC), Seoul, South Korea"],"raw_orcid":"https://orcid.org/0009-0008-9064-2862","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and Computer Science, Seoul National University, Inter-University Semiconductor Research Center (ISRC), Seoul, South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124049503","display_name":"Hyuk-Jae Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyuk-Jae Lee","raw_affiliation_strings":["Department of Electrical Engineering and Computer Science, Seoul National University, Inter-University Semiconductor Research Center (ISRC), Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-6811-9647","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and Computer Science, Seoul National University, Inter-University Semiconductor Research Center (ISRC), Seoul, South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066755055","display_name":"Chae Eun Rhee","orcid":"https://orcid.org/0000-0002-7851-1703"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Chae Eun Rhee","raw_affiliation_strings":["Department of Electronic Engineering, Hanyang University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-7851-1703","affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, Hanyang University, Seoul, South Korea","institution_ids":["https://openalex.org/I4575257"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.07854315,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"16","issue":"2","first_page":"389","last_page":"402"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.6328999996185303,"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":0.6328999996185303,"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":0.1096000000834465,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.09279999881982803,"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/scalability","display_name":"Scalability","score":0.736299991607666},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.7063999772071838},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6621999740600586},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6615999937057495},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.43689998984336853},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.4262999892234802},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.4171000123023987},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.4165000021457672},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.4097000062465668}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.800599992275238},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.736299991607666},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.7063999772071838},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6621999740600586},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6615999937057495},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5878999829292297},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.46650001406669617},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.43689998984336853},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.4262999892234802},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4171000123023987},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.4165000021457672},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.4097000062465668},{"id":"https://openalex.org/C63511323","wikidata":"https://www.wikidata.org/wiki/Q908936","display_name":"Interleaved memory","level":4,"score":0.4009999930858612},{"id":"https://openalex.org/C20574231","wikidata":"https://www.wikidata.org/wiki/Q844605","display_name":"Backward compatibility","level":2,"score":0.4000000059604645},{"id":"https://openalex.org/C2779602883","wikidata":"https://www.wikidata.org/wiki/Q15544750","display_name":"Memory architecture","level":2,"score":0.39750000834465027},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3930000066757202},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.37599998712539673},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3716000020503998},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.3668999969959259},{"id":"https://openalex.org/C123593499","wikidata":"https://www.wikidata.org/wiki/Q6008583","display_name":"In-Memory Processing","level":5,"score":0.3668000102043152},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.3499999940395355},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.33399999141693115},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.33329999446868896},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.3091000020503998},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2994000017642975},{"id":"https://openalex.org/C93446704","wikidata":"https://www.wikidata.org/wiki/Q449328","display_name":"Registered memory","level":3,"score":0.2833000123500824},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.27219998836517334},{"id":"https://openalex.org/C44274781","wikidata":"https://www.wikidata.org/wiki/Q6497132","display_name":"Lattice reduction","level":4,"score":0.2648000121116638},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C51290061","wikidata":"https://www.wikidata.org/wiki/Q1936765","display_name":"Uniform memory access","level":4,"score":0.25920000672340393},{"id":"https://openalex.org/C2778618852","wikidata":"https://www.wikidata.org/wiki/Q1128613","display_name":"Content-addressable storage","level":4,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jetcas.2026.3657823","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jetcas.2026.3657823","pdf_url":null,"source":{"id":"https://openalex.org/S142323794","display_name":"IEEE Journal on Emerging and Selected Topics in Circuits and Systems","issn_l":"2156-3357","issn":["2156-3357","2156-3365"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal on Emerging and Selected Topics in Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6757968068122864,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"embedding":[1,64,128],"layer":[2],"in":[3,28,30,93],"deep":[4],"learning":[5],"recommendation":[6,124],"models":[7],"(DLRM)":[8],"is":[9],"highly":[10],"memory-bound":[11],"and":[12,25,72,102,126,138],"exhibits":[13],"skewed,":[14],"irregular":[15],"access":[16,111],"patterns.":[17],"These":[18],"characteristics":[19],"lead":[20],"to":[21,61,135,143],"severe":[22],"load":[23],"imbalance":[24],"performance":[26],"bottlenecks":[27],"processing":[29,89],"memory":[31,58,146],"(PIM)":[32],"architectures.":[33],"We":[34],"propose":[35],"TRAM":[36,67,132],"(Two-level":[37],"Reduction":[38],"Accelerator":[39],"for":[40],"Memory),":[41],"a":[42,74],"heterogeneous":[43,145],"accelerator":[44],"that":[45,78,131],"integrates":[46],"High":[47],"Bandwidth":[48],"Memory":[49],"based":[50],"PIM":[51],"architecture":[52],"(HBM-PIM)":[53],"with":[54,85,152],"conventional":[55],"dual":[56],"in-line":[57],"modules":[59],"(DIMMs)":[60],"accelerate":[62],"batched":[63],"vector":[65],"reductions.":[66],"reduces":[68],"redundant":[69],"hot-vector":[70],"accesses":[71],"employs":[73],"host-side":[75],"scheduling":[76],"mechanism":[77],"overlaps":[79],"bank-PIM":[80],"operations":[81],"inside":[82],"DRAM":[83,155],"banks":[84],"logic-PIM":[86],"operations,":[87],"where":[88],"units":[90],"are":[91],"located":[92],"the":[94,153],"buffer":[95],"die.":[96],"This":[97],"overlap":[98],"eliminates":[99],"command-bandwidth":[100],"stalls":[101],"compute-bound":[103],"delays.":[104],"In":[105],"addition,":[106],"metadata-aware":[107],"optimizations":[108],"reduce":[109],"row/column":[110],"overhead":[112],"by":[113],"reusing":[114],"contiguous":[115],"address":[116],"patterns":[117],"within":[118],"each":[119],"bank.":[120],"Evaluation":[121],"on":[122],"six":[123],"datasets":[125],"three":[127],"dimensions":[129],"demonstrates":[130],"achieves":[133],"up":[134],"2.8\u00d7":[136],"speedup":[137],"3.0\u00d7":[139],"energy":[140],"reduction":[141],"compared":[142],"state-of-the-art":[144],"systems,":[147],"while":[148],"preserving":[149],"full":[150],"compatibility":[151],"standard":[154],"interface.":[156]},"counts_by_year":[],"updated_date":"2026-06-16T07:32:37.131356","created_date":"2026-01-28T00:00:00"}