{"id":"https://openalex.org/W4416370858","doi":"https://doi.org/10.1016/j.future.2026.108519","title":"On the energy efficiency of sparse matrix computations on multi-GPU clusters","display_name":"On the energy efficiency of sparse matrix computations on multi-GPU clusters","publication_year":2026,"publication_date":"2026-04-16","ids":{"openalex":"https://openalex.org/W4416370858","doi":"https://doi.org/10.1016/j.future.2026.108519"},"language":"en","primary_location":{"id":"doi:10.1016/j.future.2026.108519","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108519","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1016/j.future.2026.108519","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5006496977","display_name":"Massimo Bernaschi","orcid":"https://orcid.org/0000-0003-3661-9836"},"institutions":[{"id":"https://openalex.org/I4210155236","display_name":"National Research Council","ror":"https://ror.org/04zaypm56","country_code":"IT","type":"nonprofit","lineage":["https://openalex.org/I4210155236"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Massimo Bernaschi","raw_affiliation_strings":["Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy"],"raw_orcid":"https://orcid.org/0000-0003-3661-9836","affiliations":[{"raw_affiliation_string":"Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy","institution_ids":["https://openalex.org/I4210155236"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005168115","display_name":"Alessandro Celestini","orcid":"https://orcid.org/0000-0002-8493-4550"},"institutions":[{"id":"https://openalex.org/I4210155236","display_name":"National Research Council","ror":"https://ror.org/04zaypm56","country_code":"IT","type":"nonprofit","lineage":["https://openalex.org/I4210155236"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Alessandro Celestini","raw_affiliation_strings":["Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy"],"raw_orcid":"https://orcid.org/0000-0002-8493-4550","affiliations":[{"raw_affiliation_string":"Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy","institution_ids":["https://openalex.org/I4210155236"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Giorgio Richelli","orcid":null},"institutions":[{"id":"https://openalex.org/I4210155236","display_name":"National Research Council","ror":"https://ror.org/04zaypm56","country_code":"IT","type":"nonprofit","lineage":["https://openalex.org/I4210155236"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Giorgio Richelli","raw_affiliation_strings":["Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via dei Taurini 19, 00185, Rome, Italy","institution_ids":["https://openalex.org/I4210155236"]}]},{"author_position":"last","author":{"id":null,"display_name":"Pasqua D\u2019Ambra","orcid":"https://orcid.org/0000-0003-2047-4986"},"institutions":[{"id":"https://openalex.org/I4210129308","display_name":"Institute of Protein Biochemistry","ror":"https://ror.org/03eqeqg74","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210129308","https://openalex.org/I4210155236"]},{"id":"https://openalex.org/I4210155236","display_name":"National Research Council","ror":"https://ror.org/04zaypm56","country_code":"IT","type":"nonprofit","lineage":["https://openalex.org/I4210155236"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Pasqua D\u2019Ambra","raw_affiliation_strings":["Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via Pietro Castellino, 111, 80131, Naples, Italy"],"raw_orcid":"https://orcid.org/0000-0003-2047-4986","affiliations":[{"raw_affiliation_string":"Institute for Applied Computing \u201dMauro Picone\u201d, National Research Council, Via Pietro Castellino, 111, 80131, Naples, Italy","institution_ids":["https://openalex.org/I4210129308","https://openalex.org/I4210155236"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5005168115"],"corresponding_institution_ids":["https://openalex.org/I4210155236"],"apc_list":{"value":3340,"currency":"USD","value_usd":3340},"apc_paid":{"value":3340,"currency":"USD","value_usd":3340},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00956142,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"183","issue":null,"first_page":"108519","last_page":"108519"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8920000195503235,"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":0.8920000195503235,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.033799998462200165,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.01759999990463257,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.7325999736785889},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6654000282287598},{"id":"https://openalex.org/keywords/graphics-processing-unit","display_name":"Graphics processing unit","score":0.5870000123977661},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5827999711036682},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.5462999939918518},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.5148000121116638},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.5066999793052673},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.48840001225471497},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.45680001378059387},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.4381999969482422}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8140000104904175},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.7325999736785889},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6654000282287598},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6298999786376953},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.5870000123977661},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5827999711036682},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.5777999758720398},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.5462999939918518},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.5148000121116638},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.5073999762535095},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.5066999793052673},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.48840001225471497},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.45680001378059387},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.4381999969482422},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.427700012922287},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.4212999939918518},{"id":"https://openalex.org/C79470037","wikidata":"https://www.wikidata.org/wiki/Q279748","display_name":"Out-of-core algorithm","level":2,"score":0.41929998993873596},{"id":"https://openalex.org/C76518257","wikidata":"https://www.wikidata.org/wiki/Q271680","display_name":"Software framework","level":5,"score":0.38670000433921814},{"id":"https://openalex.org/C11644782","wikidata":"https://www.wikidata.org/wiki/Q15401790","display_name":"Cost efficiency","level":2,"score":0.365200012922287},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.34439998865127563},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.32409998774528503},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3125},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.31139999628067017},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C139352143","wikidata":"https://www.wikidata.org/wiki/Q82571","display_name":"Linear algebra","level":2,"score":0.29910001158714294},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.2921000123023987},{"id":"https://openalex.org/C106515295","wikidata":"https://www.wikidata.org/wiki/Q26806595","display_name":"Parallel processing","level":2,"score":0.29100000858306885},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.28790000081062317},{"id":"https://openalex.org/C120373497","wikidata":"https://www.wikidata.org/wiki/Q1087987","display_name":"Parallel algorithm","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.267300009727478},{"id":"https://openalex.org/C91481028","wikidata":"https://www.wikidata.org/wiki/Q1054686","display_name":"Distributed memory","level":3,"score":0.26100000739097595},{"id":"https://openalex.org/C2780365336","wikidata":"https://www.wikidata.org/wiki/Q25047934","display_name":"Single-core","level":2,"score":0.25839999318122864},{"id":"https://openalex.org/C41045048","wikidata":"https://www.wikidata.org/wiki/Q202843","display_name":"Linear programming","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1016/j.future.2026.108519","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108519","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2510.02878","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2510.02878","pdf_url":"https://arxiv.org/pdf/2510.02878","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.02878","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.02878","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1016/j.future.2026.108519","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108519","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6517815977","display_name":"FOSTERING THE EUROPEAN ENERGY TRANSITION WITH EXASCALE","funder_award_id":"101144014","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320311048","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479"},{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F5497039910","display_name":"Ministero dell'Istruzione e del Merito","ror":"https://ror.org/01ehyh486"},{"id":"https://openalex.org/F7311471023","display_name":"NextGenerationEU","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,121],"investigate":[1],"the":[2,50,83,113,133,139,167],"energy":[3,109,130,161],"efficiency":[4,86],"of":[5,53,68,92,117,132],"a":[6,54,65],"library":[7,16,168],"designed":[8],"for":[9,74,127],"parallel":[10,37],"computations":[11,147],"with":[12],"sparse":[13,43],"matrices.":[14],"The":[15],"leverages":[17],"high-performance,":[18],"energy-efficient":[19],"Graphics":[20],"Processing":[21],"Unit":[22],"(GPU)":[23],"accelerators":[24],"to":[25,35],"enable":[26],"large-scale":[27,88],"scientific":[28],"applications.":[29],"Our":[30,141],"primary":[31],"development":[32],"objective":[33],"was":[34],"maximize":[36],"performance":[38,85],"and":[39,125,137,148,154,160],"scalability":[40],"in":[41],"solving":[42],"linear":[44],"systems":[45,89],"whose":[46],"dimensions":[47],"far":[48],"exceed":[49],"memory":[51,153],"capacity":[52],"single":[55],"node.":[56],"To":[57],"this":[58,101],"end,":[59],"we":[60,103,164],"devised":[61],"methods":[62],"that":[63,111,144,166],"expose":[64],"high":[66],"degree":[67],"parallelism":[69],"while":[70],"optimizing":[71,145],"algorithmic":[72],"implementations":[73],"efficient":[75],"multi-GPU":[76],"usage.":[77],"Previous":[78],"work":[79],"has":[80],"already":[81],"demonstrated":[82],"library\u2019s":[84,134],"on":[87,176],"comprising":[90],"thousands":[91],"NVIDIA":[93],"GPUs,":[94],"achieving":[95],"improvements":[96],"over":[97,172],"state-of-the-art":[98],"solutions.":[99],"In":[100],"paper,":[102],"extend":[104],"those":[105],"results":[106,142],"by":[107],"providing":[108],"profiles":[110],"address":[112],"growing":[114],"sustainability":[115],"requirements":[116],"modern":[118],"HPC":[119],"platforms.":[120],"present":[122],"our":[123],"methodology":[124],"tools":[126],"accurate":[128],"runtime":[129],"measurements":[131],"core":[135],"components":[136],"discuss":[138],"findings.":[140],"confirm":[143],"GPU":[146],"minimizing":[149],"data":[150],"movement":[151],"across":[152],"computing":[155],"nodes":[156],"reduces":[157],"both":[158],"time-to-solution":[159],"consumption.":[162],"Moreover,":[163],"show":[165],"delivers":[169],"substantial":[170],"advantages":[171],"comparable":[173],"software":[174],"frameworks":[175],"standard":[177],"benchmarks.":[178]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-10-10T00:00:00"}