{"id":"https://openalex.org/W7166737321","doi":"https://doi.org/10.1145/3774895.3815549","title":"An Energy Characterization Study of ML operators in XNNPACK on RISC-V Vector Hardware","display_name":"An Energy Characterization Study of ML operators in XNNPACK on RISC-V Vector Hardware","publication_year":2026,"publication_date":"2026-06-30","ids":{"openalex":"https://openalex.org/W7166737321","doi":"https://doi.org/10.1145/3774895.3815549"},"language":null,"primary_location":{"id":"doi:10.1145/3774895.3815549","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774895.3815549","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 40th ACM International Conference on Supercomputing - Workshops","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3774895.3815549","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124448726","display_name":"Ruimin Shi","orcid":null},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Ruimin Shi","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0009-0003-4387-367X","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114438988","display_name":"Samuel Miksits","orcid":null},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Samuel Miksits","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0009-0008-4601-069X","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074163062","display_name":"Ronald G. Minnich","orcid":"https://orcid.org/0009-0003-1777-562X"},"institutions":[{"id":"https://openalex.org/I4210122178","display_name":"Hewlett Packard Enterprise (United States)","ror":"https://ror.org/020x0c621","country_code":"US","type":"company","lineage":["https://openalex.org/I4210122178"]},{"id":"https://openalex.org/I4210163432","display_name":"Hewlett Packard Enterprise (Ireland)","ror":"https://ror.org/05cc0x492","country_code":"IE","type":"company","lineage":["https://openalex.org/I4210163432"]}],"countries":["IE","US"],"is_corresponding":false,"raw_author_name":"Ron Minnich","raw_affiliation_strings":["Hewlett Packard Enterprise, Livermore, USA"],"raw_orcid":"https://orcid.org/0009-0003-1777-562X","affiliations":[{"raw_affiliation_string":"Hewlett Packard Enterprise, Livermore, USA","institution_ids":["https://openalex.org/I4210122178","https://openalex.org/I4210163432"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043476271","display_name":"Maya Gokhale","orcid":"https://orcid.org/0000-0003-4229-5735"},"institutions":[{"id":"https://openalex.org/I1282311441","display_name":"Lawrence Livermore National Laboratory","ror":"https://ror.org/041nk4h53","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282311441","https://openalex.org/I1330989302","https://openalex.org/I198811213","https://openalex.org/I4210138311"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Maya Gokhale","raw_affiliation_strings":["Lawrence Livermore National Laboratory, Livermore, USA"],"raw_orcid":"https://orcid.org/0000-0003-4229-5735","affiliations":[{"raw_affiliation_string":"Lawrence Livermore National Laboratory, Livermore, USA","institution_ids":["https://openalex.org/I1282311441"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037069204","display_name":"Ivy Peng","orcid":"https://orcid.org/0000-0003-4158-3583"},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Ivy Peng","raw_affiliation_strings":["KTH Royal Institute of Technology, Stockholm, Sweden"],"raw_orcid":"https://orcid.org/0000-0003-4158-3583","affiliations":[{"raw_affiliation_string":"KTH Royal Institute of Technology, Stockholm, Sweden","institution_ids":["https://openalex.org/I86987016"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"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":0.85501946,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"154","last_page":"159"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.7480999827384949,"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":0.7480999827384949,"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/T14347","display_name":"Big Data and Digital Economy","score":0.044599998742341995,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.02930000051856041,"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/speedup","display_name":"Speedup","score":0.7307999730110168},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.5278000235557556},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.49540001153945923},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.42329999804496765},{"id":"https://openalex.org/keywords/extension","display_name":"Extension (predicate logic)","score":0.40880000591278076},{"id":"https://openalex.org/keywords/support-vector-machine","display_name":"Support vector machine","score":0.38040000200271606}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7307999730110168},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.670799970626831},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.5278000235557556},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.49540001153945923},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.45809999108314514},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.42329999804496765},{"id":"https://openalex.org/C2778029271","wikidata":"https://www.wikidata.org/wiki/Q5421931","display_name":"Extension (predicate logic)","level":2,"score":0.40880000591278076},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.38040000200271606},{"id":"https://openalex.org/C65232700","wikidata":"https://www.wikidata.org/wiki/Q5656403","display_name":"Hardware architecture","level":3,"score":0.36820000410079956},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.33739998936653137},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.33160001039505005},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.32350000739097595},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3165000081062317},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.28600001335144043},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27379998564720154},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2718999981880188},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C2985973956","wikidata":"https://www.wikidata.org/wiki/Q1617745","display_name":"High energy","level":2,"score":0.25279998779296875}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3774895.3815549","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774895.3815549","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 40th ACM International Conference on Supercomputing - Workshops","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3774895.3815549","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774895.3815549","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 40th ACM International Conference on Supercomputing - Workshops","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8962506055831909}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":6,"referenced_works":["https://openalex.org/W4394805411","https://openalex.org/W4411345920","https://openalex.org/W4412009493","https://openalex.org/W4416003937","https://openalex.org/W4416004048","https://openalex.org/W4416046488"],"related_works":[],"abstract_inverted_index":{"The":[0],"RISC-V":[1,58],"Vector":[2],"Extension":[3],"(RVV)":[4],"is":[5],"gaining":[6],"traction":[7],"as":[8],"an":[9],"open-model":[10],"architecture":[11],"for":[12,121],"machine":[13,22],"learning":[14,23],"workloads.":[15],"However,":[16],"the":[17,76,90,94,113],"energy":[18,82,103],"characteristics":[19],"of":[20,39,78],"fundamental":[21],"operations":[24,45],"on":[25,55,80],"RVV":[26,53,79,95],"hardware":[27],"remain":[28],"largely":[29],"unstudied.":[30],"In":[31],"this":[32],"work,":[33],"we":[34],"conduct":[35],"a":[36,56,118],"preliminary":[37],"study":[38],"highly":[40],"optimized":[41],"micro":[42],"element-wise":[43],"ML":[44,123],"and":[46,70,83,88,105],"representative":[47],"model-level":[48],"workloads":[49],"executed":[50],"by":[51],"XNNPACK\u2019s":[52],"backend":[54,96],"real":[57],"board,":[59],"BPI-F3":[60],"with":[61],"256-bit":[62],"VLEN":[63],"RVV,":[64],"across":[65],"six":[66],"data":[67],"types":[68],"(qs8/qu8/bf16/f16/f32/s32)":[69],"four":[71],"LMUL":[72],"selections.":[73],"We":[74],"quantify":[75],"impact":[77],"power,":[81],"energy-per-element.":[84],"On":[85],"both":[86],"single":[87],"multi-cores,":[89],"results":[91],"show":[92],"that":[93],"achieves":[97],"1.41":[98],"\u2212":[99],"4.78":[100],"\u00d7":[101,109],"higher":[102],"efficiency":[104],"up":[106],"to":[107],"5.38":[108],"performance":[110],"speedup":[111],"than":[112],"non-vectorized":[114],"backend,":[115],"making":[116],"it":[117],"promising":[119],"candidate":[120],"energy-constrained":[122],"inference":[124],"deployment.":[125]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-01T00:00:00"}