{"id":"https://openalex.org/W4412610665","doi":"https://doi.org/10.1109/hpcc64274.2024.00039","title":"OBSD: On-The-Fly Block-Wise Sparse Distillation Accelerating SpGEMMs in DNN Applications","display_name":"OBSD: On-The-Fly Block-Wise Sparse Distillation Accelerating SpGEMMs in DNN Applications","publication_year":2024,"publication_date":"2024-12-13","ids":{"openalex":"https://openalex.org/W4412610665","doi":"https://doi.org/10.1109/hpcc64274.2024.00039"},"language":"en","primary_location":{"id":"doi:10.1109/hpcc64274.2024.00039","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpcc64274.2024.00039","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on High Performance Computing and Communications (HPCC)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5103111619","display_name":"Yanhuan Liu","orcid":"https://orcid.org/0009-0001-8460-8173"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanhuan Liu","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101630070","display_name":"Wenming Li","orcid":"https://orcid.org/0000-0003-4069-2251"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenming Li","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014918710","display_name":"Kunming Zhang","orcid":"https://orcid.org/0000-0001-7088-8154"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kunming Zhang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088057715","display_name":"Zhihua Fan","orcid":"https://orcid.org/0000-0002-5950-7370"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihua Fan","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101474129","display_name":"Haibin Wu","orcid":"https://orcid.org/0009-0000-2949-2297"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haibin Wu","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034315729","display_name":"Lexin Wang","orcid":"https://orcid.org/0000-0002-1136-063X"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lexin Wang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100379419","display_name":"Tianyu Liu","orcid":"https://orcid.org/0000-0001-8107-8587"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianyu Liu","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015955856","display_name":"Yuhang Qiu","orcid":"https://orcid.org/0000-0002-9706-4013"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhang Qiu","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101970243","display_name":"Zhen Wang","orcid":"https://orcid.org/0009-0003-8853-9915"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhen Wang","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023098180","display_name":"Xiaochun Ye","orcid":"https://orcid.org/0000-0003-4598-1685"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaochun Ye","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011407484","display_name":"Dongrui Fan","orcid":"https://orcid.org/0000-0001-5219-0908"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongrui Fan","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101472673","display_name":"Xuejun An","orcid":"https://orcid.org/0009-0005-0494-6332"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuejun An","raw_affiliation_strings":["Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, CAS,State Key Lab of Processors,Beijing,China","institution_ids":["https://openalex.org/I4210090176"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210090176"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.43852968,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"224","last_page":"231"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9829000234603882,"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"}},"topics":[{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9829000234603882,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9819999933242798,"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/T13553","display_name":"Age of Information Optimization","score":0.9710999727249146,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.7552657127380371},{"id":"https://openalex.org/keywords/on-the-fly","display_name":"On the fly","score":0.7408313155174255},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6413122415542603},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.6128594279289246},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3688545227050781},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.18582120537757874},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.06912988424301147},{"id":"https://openalex.org/keywords/combinatorics","display_name":"Combinatorics","score":0.061270713806152344},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.060507357120513916},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.05761405825614929}],"concepts":[{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.7552657127380371},{"id":"https://openalex.org/C2781020372","wikidata":"https://www.wikidata.org/wiki/Q533093","display_name":"On the fly","level":2,"score":0.7408313155174255},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6413122415542603},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.6128594279289246},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3688545227050781},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.18582120537757874},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.06912988424301147},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.061270713806152344},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.060507357120513916},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.05761405825614929}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpcc64274.2024.00039","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpcc64274.2024.00039","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on High Performance Computing and Communications (HPCC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320326444","display_name":"Nova","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2515287984","https://openalex.org/W2516141709","https://openalex.org/W2883542588","https://openalex.org/W2904902077","https://openalex.org/W2912776959","https://openalex.org/W2945146780","https://openalex.org/W2953212265","https://openalex.org/W2979310060","https://openalex.org/W2979439447","https://openalex.org/W2980104813","https://openalex.org/W3016542674","https://openalex.org/W3016769527","https://openalex.org/W3040024858","https://openalex.org/W3130554079","https://openalex.org/W3187481008","https://openalex.org/W3187908937","https://openalex.org/W4224919827","https://openalex.org/W4236868170","https://openalex.org/W4281800613","https://openalex.org/W4286308246","https://openalex.org/W4324292875","https://openalex.org/W4360831848","https://openalex.org/W4386106826","https://openalex.org/W4389724193","https://openalex.org/W4391664585"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"The":[0,167],"mainstream":[1],"deep":[2],"neural":[3],"networks":[4],"(DNNs)":[5],"widely":[6],"adopt":[7],"pruning":[8],"techniques":[9],"to":[10,58,143,181],"alleviate":[11],"network":[12],"overfitting":[13],"and":[14,20,77,104,122,164,201],"computational":[15],"complexity.":[16],"Simultaneously,":[17],"the":[18,45,53,69,74,82,86,116,129,156,208],"weight":[19,121],"activation":[21,123],"data":[22,130,147,150],"exhibit":[23],"increasingly":[24],"noticeable":[25],"block-wise":[26,109,117,162,165],"sparsity.":[27],"Current":[28],"DNN-specific":[29],"accelerators":[30],"mainly":[31],"focus":[32],"on":[33,73,81,193],"element-wise":[34],"while":[35],"neglecting":[36],"sparse":[37],"features":[38],"of":[39,44,62,85,146,176,205],"block":[40],"granularity.":[41],"Meanwhile,":[42],"one":[43],"state-of-the-art":[46],"work,":[47,91],"HIRAC,":[48],"combining":[49],"matrix":[50],"tiling":[51],"with":[52,154,217],"fast":[54],"packing":[55],"algorithm,":[56],"SorPack,":[57],"achieve":[59],"effective":[60],"acceleration":[61],"Sparse":[63],"General":[64],"Matrix":[65],"Multiplications":[66],"(SpGEMMs).":[67],"However,":[68],"SorPack":[70],"algorithm":[71],"executes":[72],"host":[75],"CPU,":[76],"its":[78],"runtime":[79],"lies":[80],"critical":[83],"path":[84],"overall":[87],"execution.":[88],"In":[89],"this":[90],"a":[92,138,194,198,213],"specific":[93],"SpGEMM":[94],"accelerator":[95],"named":[96],"OBSD":[97,173,211],"is":[98,112,126,141],"proposed":[99,113],"which":[100,125],"achieves":[101,174],"high":[102],"performance":[103],"energy":[105,202],"efficiency.":[106],"An":[107,190],"on-the-fly":[108],"distillation":[110],"approach":[111],"for":[114,158,183],"leveraging":[115],"sparsity":[118],"in":[119,186],"both":[120],"data,":[124],"implemented":[127],"during":[128],"loading":[131],"process,":[132],"consuming":[133],"minimal":[134],"additional":[135],"time.":[136],"Moreover,":[137],"data-flow":[139],"architecture":[140],"devised":[142],"improve":[144],"efficiency":[145,203],"exchange":[148],"among":[149],"processing":[151],"elements":[152],"(DPs)":[153],"investigating":[155],"relationship":[157],"different":[159],"partition":[160],"sizes,":[161],"sparsity,":[163],"distance.":[166],"evaluation":[168,192],"results":[169],"demonstrate":[170],"that:":[171],"(1)":[172],"average":[175],"1.89\u00d7":[177],"speedup":[178,200],"as":[179],"compared":[180],"HIRAC":[182],"representative":[184],"matrices":[185],"DNN":[187,195],"workloads.":[188],"(2)":[189],"end-to-end":[191],"model":[196],"shows":[197],"2.41\u00d7":[199],"improvement":[204],"1.99\u00d7":[206],"over":[207],"HIRAC.":[209],"(3)":[210],"possesses":[212],"power":[214],"consumption":[215],"12.9W":[216],"56.6mm2area":[218],"@28nm":[219],"TSMC.":[220]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}
