{"id":"https://openalex.org/W4360831848","doi":"https://doi.org/10.1109/hpca56546.2023.10070977","title":"HIRAC: A Hierarchical Accelerator with Sorting-based Packing for SpGEMMs in DNN Applications","display_name":"HIRAC: A Hierarchical Accelerator with Sorting-based Packing for SpGEMMs in DNN Applications","publication_year":2023,"publication_date":"2023-02-01","ids":{"openalex":"https://openalex.org/W4360831848","doi":"https://doi.org/10.1109/hpca56546.2023.10070977"},"language":"en","primary_location":{"id":"doi:10.1109/hpca56546.2023.10070977","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10070977","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","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/A5087978342","display_name":"Hesam Shabani","orcid":null},"institutions":[{"id":"https://openalex.org/I186143895","display_name":"Lehigh University","ror":"https://ror.org/012afjb06","country_code":"US","type":"education","lineage":["https://openalex.org/I186143895"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Hesam Shabani","raw_affiliation_strings":["Lehigh University,USA","Lehigh University, USA"],"affiliations":[{"raw_affiliation_string":"Lehigh University,USA","institution_ids":["https://openalex.org/I186143895"]},{"raw_affiliation_string":"Lehigh University, USA","institution_ids":["https://openalex.org/I186143895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017789178","display_name":"Abhishek Singh","orcid":"https://orcid.org/0000-0002-4533-6347"},"institutions":[{"id":"https://openalex.org/I186143895","display_name":"Lehigh University","ror":"https://ror.org/012afjb06","country_code":"US","type":"education","lineage":["https://openalex.org/I186143895"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abhishek Singh","raw_affiliation_strings":["Lehigh University,USA","Lehigh University, USA"],"affiliations":[{"raw_affiliation_string":"Lehigh University,USA","institution_ids":["https://openalex.org/I186143895"]},{"raw_affiliation_string":"Lehigh University, USA","institution_ids":["https://openalex.org/I186143895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072936202","display_name":"Bishoy Youhana","orcid":null},"institutions":[{"id":"https://openalex.org/I186143895","display_name":"Lehigh University","ror":"https://ror.org/012afjb06","country_code":"US","type":"education","lineage":["https://openalex.org/I186143895"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bishoy Youhana","raw_affiliation_strings":["Lehigh University,USA","Lehigh University, USA"],"affiliations":[{"raw_affiliation_string":"Lehigh University,USA","institution_ids":["https://openalex.org/I186143895"]},{"raw_affiliation_string":"Lehigh University, USA","institution_ids":["https://openalex.org/I186143895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028757963","display_name":"Xiaochen Guo","orcid":"https://orcid.org/0000-0001-7704-0412"},"institutions":[{"id":"https://openalex.org/I186143895","display_name":"Lehigh University","ror":"https://ror.org/012afjb06","country_code":"US","type":"education","lineage":["https://openalex.org/I186143895"]},{"id":"https://openalex.org/I4210087596","display_name":"Qualcomm (United States)","ror":"https://ror.org/002zrf773","country_code":"US","type":"company","lineage":["https://openalex.org/I4210087596"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaochen Guo","raw_affiliation_strings":["Lehigh University,USA","Lehigh University, USA","Qualcomm, Inc., USA"],"affiliations":[{"raw_affiliation_string":"Lehigh University,USA","institution_ids":["https://openalex.org/I186143895"]},{"raw_affiliation_string":"Lehigh University, USA","institution_ids":["https://openalex.org/I186143895"]},{"raw_affiliation_string":"Qualcomm, Inc., USA","institution_ids":["https://openalex.org/I4210087596"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5087978342"],"corresponding_institution_ids":["https://openalex.org/I186143895"],"apc_list":null,"apc_paid":null,"fwci":4.0358,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.94336147,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"247","last_page":"258"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9951000213623047,"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":0.9951000213623047,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9937000274658203,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.9905999898910522,"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":0.7768545150756836},{"id":"https://openalex.org/keywords/interconnection","display_name":"Interconnection","score":0.658038854598999},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6094810366630554},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5959973335266113},{"id":"https://openalex.org/keywords/sorting","display_name":"Sorting","score":0.5549491047859192},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5109683275222778},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.49042245745658875},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.48174428939819336},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4512082040309906},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.4350050091743469},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.374755859375},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.315651535987854},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.1508621871471405}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7768545150756836},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.658038854598999},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6094810366630554},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5959973335266113},{"id":"https://openalex.org/C111696304","wikidata":"https://www.wikidata.org/wiki/Q2303697","display_name":"Sorting","level":2,"score":0.5549491047859192},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5109683275222778},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.49042245745658875},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.48174428939819336},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4512082040309906},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.4350050091743469},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.374755859375},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.315651535987854},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1508621871471405},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"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/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca56546.2023.10070977","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10070977","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8899999856948853}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W1996274599","https://openalex.org/W2009654791","https://openalex.org/W2013638793","https://openalex.org/W2118231264","https://openalex.org/W2285660444","https://openalex.org/W2606722458","https://openalex.org/W2725159389","https://openalex.org/W2896409484","https://openalex.org/W2945146780","https://openalex.org/W2952294252","https://openalex.org/W2963594949","https://openalex.org/W2980104813","https://openalex.org/W3016542674","https://openalex.org/W3016832937","https://openalex.org/W3024621361","https://openalex.org/W3040024858","https://openalex.org/W3103168911","https://openalex.org/W3155922894","https://openalex.org/W3158831985","https://openalex.org/W3167841307","https://openalex.org/W3187908937","https://openalex.org/W3190761184","https://openalex.org/W3206453033","https://openalex.org/W4224612674","https://openalex.org/W4247340992","https://openalex.org/W4293024984","https://openalex.org/W4293084046","https://openalex.org/W6755843862","https://openalex.org/W6765111094","https://openalex.org/W6796114172"],"related_works":["https://openalex.org/W4287593139","https://openalex.org/W3099313426","https://openalex.org/W1506547947","https://openalex.org/W2086146848","https://openalex.org/W4206811032","https://openalex.org/W2995605830","https://openalex.org/W4239424132","https://openalex.org/W2596457687","https://openalex.org/W3212757063","https://openalex.org/W2086123442"],"abstract_inverted_index":{"The":[0,131,150,203,220],"state-of-the-art":[1,257],"deep":[2],"neural":[3],"network":[4,215],"(DNN)":[5],"models":[6],"use":[7,176],"pruning":[8],"to":[9,20,58,101,118,135,153,160,165,191,226,255,277],"avoid":[10,59,174],"over-fitting":[11],"and":[12,23,31,92,138,173,213,231,270],"reduce":[13],"the":[14,46,145,155,175,198,201,224,237,256,292],"number":[15,146],"of":[16,49,54,73,147,177,200,207,244,251],"parameters.":[17],"In":[18,181,262],"order":[19],"improve":[21,228],"storage":[22],"computational":[24],"efficiency,":[25],"only":[26],"nonzero":[27,148],"elements":[28,62],"are":[29,34],"stored,":[30],"their":[32],"locations":[33],"encoded":[35],"into":[36,123],"a":[37,97,120,124,183,193,208,248,266,271,283],"sparse":[38,121,258],"format.":[39],"Sparse":[40],"General":[41],"Matrix":[42],"Multiplication":[43],"(SpGEMM)":[44],"is":[45,57,116,134,152,189],"kernel":[47],"computation":[48],"DNN-based":[50],"applications.":[51,219],"One":[52],"challenge":[53,82],"computing":[55],"SpGEMM":[56,103],"multiplying":[60],"zero":[61],"while":[63],"keeping":[64],"hardware":[65,69,229],"utilization":[66,230],"high":[67,90],"in":[68],"accelerators":[70],"that":[71,127,158,196],"consist":[72],"processing":[74],"element":[75],"(PE)":[76],"arrays.":[77],"Prior":[78],"work":[79,95],"tackling":[80],"this":[81],"typically":[83],"requires":[84],"complex":[85,107,178],"interconnection":[86,108,179,214],"networks,":[87],"which":[88],"adds":[89],"area":[91,268],"energy":[93],"costs.This":[94],"proposes":[96],"HW/SW":[98],"co-design":[99],"architecture":[100,205],"compute":[102],"efficiently":[104],"without":[105],"requiring":[106],"networks.":[109,180],"A":[110],"novel":[111,209],"fast":[112],"packing":[113],"algorithm,":[114],"SorPack,":[115],"proposed":[117,190],"convert":[119],"matrix":[122,126],"dense":[125],"increases":[128],"PE":[129,210],"utilization.":[130],"key":[132],"idea":[133],"sort":[136],"columns":[137],"rows":[139],"inside":[140],"each":[141,166],"submatrix":[142],"based":[143],"on":[144,236,247,282],"elements.":[149],"goal":[151],"keep":[154],"partial":[156],"sums":[157],"need":[159],"be":[161,170],"added":[162,171],"together":[163],"close":[164],"other,":[167],"hence":[168],"can":[169],"locally":[172],"addition,":[182,263],"new":[184],"tile-based":[185],"hierarchical":[186],"architecture,":[187],"HIRAC,":[188],"provide":[192],"scalable":[194],"system":[195,233],"maximizes":[197],"parallelism":[199],"PEs.":[202],"HIRAC":[204,225,240,264],"consists":[206],"array":[211],"design":[212],"tailored":[216],"for":[217],"DNN":[218,252,259,284],"SorPack":[221],"algorithm":[222],"complements":[223],"further":[227],"overall":[232],"performance.":[234],"Based":[235],"evaluation":[238,281],"results,":[239],"achieves":[241],"an":[242,287],"average":[243],"3.2\u00d7":[245],"speedup":[246],"single":[249],"layer":[250],"as":[253,275],"compared":[254,276],"accelerator":[260],"SIGMA.":[261,278],"has":[265],"9.5%":[267],"reduction":[269,274,290],"32%":[272],"power":[273],"An":[279],"end-to-end":[280],"model":[285],"shows":[286],"8.2\u00d7":[288],"runtime":[289],"over":[291],"TPU.":[293]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":2}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}