{"id":"https://openalex.org/W4415380946","doi":"https://doi.org/10.21203/rs.3.rs-7890483/v1","title":"Racing to Idle: Energy Efficiency of Matrix Multiplication on Heterogeneous CPU and GPU Architectures","display_name":"Racing to Idle: Energy Efficiency of Matrix Multiplication on Heterogeneous CPU and GPU Architectures","publication_year":2025,"publication_date":"2025-10-21","ids":{"openalex":"https://openalex.org/W4415380946","doi":"https://doi.org/10.21203/rs.3.rs-7890483/v1"},"language":"en","primary_location":{"id":"doi:10.21203/rs.3.rs-7890483/v1","doi":"https://doi.org/10.21203/rs.3.rs-7890483/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-7890483/v1","pdf_url":"https://www.researchsquare.com/article/rs-7890483/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Research Square","raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.researchsquare.com/article/rs-7890483/latest.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113728039","display_name":"Mufakir Qamar Ansari","orcid":"https://orcid.org/0009-0000-8719-7456","observed_orcids":["https://orcid.org/0009-0000-8719-7456"]},"institutions":[{"id":"https://openalex.org/I90871651","display_name":"University of Toledo","ror":"https://ror.org/01pbdzh19","country_code":"US","type":"education","lineage":["https://openalex.org/I90871651"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Mufakir Qamar Ansari","raw_affiliation_strings":["The University of Toledo"],"raw_orcid":"https://orcid.org/0009-0000-8719-7456","affiliations":[{"raw_affiliation_string":"The University of Toledo","institution_ids":["https://openalex.org/I90871651"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5113728039"],"corresponding_institution_ids":["https://openalex.org/I90871651"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994999766349792,"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":0.9994999766349792,"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/T14098","display_name":"Energy Efficiency in Computing","score":9.999999747378752e-05,"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/T10363","display_name":"Low-power high-performance VLSI design","score":9.999999747378752e-05,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/heterogeneous-computing","display_name":"heterogeneous computing","score":0.7900000214576721},{"id":"https://openalex.org/keywords/energy-efficiency","display_name":"energy efficiency","score":0.8989999890327454},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"matrix multiplication","score":0.9589999914169312},{"id":"https://openalex.org/keywords/cpu-vs-gpu-performance-comparison","display_name":"CPU vs GPU performance comparison","score":0.8140000104904175},{"id":"https://openalex.org/keywords/power-consumption","display_name":"power consumption","score":0.5770000219345093},{"id":"https://openalex.org/keywords/integrated-gpu","display_name":"integrated GPU","score":0.9950000047683716},{"id":"https://openalex.org/keywords/multi-core-processors","display_name":"multi-core processors","score":0.7070000171661377},{"id":"https://openalex.org/keywords/laptop-hardware","display_name":"laptop hardware","score":0.4950000047683716},{"id":"https://openalex.org/keywords/energy-aware-computing","display_name":"energy-aware computing","score":0.6380000114440918}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7627000212669373},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.7174000144004822},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.7009999752044678},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6660000085830688},{"id":"https://openalex.org/C16320812","wikidata":"https://www.wikidata.org/wiki/Q1812200","display_name":"Idle","level":2,"score":0.6363999843597412},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.5616000294685364},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.48069998621940613},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.47699999809265137},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.4429999887943268},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4401000142097473},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.43389999866485596},{"id":"https://openalex.org/C118993495","wikidata":"https://www.wikidata.org/wiki/Q5042828","display_name":"Electrical efficiency","level":3,"score":0.4075999855995178},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.37450000643730164},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3531999886035919},{"id":"https://openalex.org/C520301825","wikidata":"https://www.wikidata.org/wiki/Q380170","display_name":"Energy conservation","level":2,"score":0.35019999742507935},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3082999885082245},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.3070000112056732},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.30630001425743103},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2935999929904938},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2915000021457672},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.2897999882698059},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2734000086784363},{"id":"https://openalex.org/C172430144","wikidata":"https://www.wikidata.org/wiki/Q17111997","display_name":"Symmetric multiprocessor system","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C35912277","wikidata":"https://www.wikidata.org/wiki/Q1243369","display_name":"Double-precision floating-point format","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.25220000743865967}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.21203/rs.3.rs-7890483/v1","doi":"https://doi.org/10.21203/rs.3.rs-7890483/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-7890483/v1","pdf_url":"https://www.researchsquare.com/article/rs-7890483/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Research Square","raw_type":"posted-content"},{"id":"pmh:oai:arXiv.org:2507.20063","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.20063","pdf_url":"https://arxiv.org/pdf/2507.20063","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2507.20063","doi":"https://doi.org/10.48550/arxiv.2507.20063","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2507.20063","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,"listed_in":[],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.21203/rs.3.rs-7890483/v1","doi":"https://doi.org/10.21203/rs.3.rs-7890483/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-7890483/v1","pdf_url":"https://www.researchsquare.com/article/rs-7890483/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Research Square","raw_type":"posted-content"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.571399986743927}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415380946.pdf","grobid_xml":"https://content.openalex.org/works/W4415380946.grobid-xml"},"referenced_works_count":3,"referenced_works":["https://openalex.org/W4378942706","https://openalex.org/W4417267327","https://openalex.org/W7126101480"],"related_works":["https://openalex.org/W2038122178","https://openalex.org/W2122407197","https://openalex.org/W7124680658","https://openalex.org/W7167263308","https://openalex.org/W2412759392","https://openalex.org/W7221412785","https://openalex.org/W4417267327","https://openalex.org/W2770891419","https://openalex.org/W4416370858","https://openalex.org/W1985364901"],"abstract_inverted_index":{"Abstract":[0],"Heterogeneous":[1],"computing":[2],"has":[3],"emerged":[4],"as":[5],"an":[6,83],"essential":[7],"approach":[8],"to":[9,165,181],"overcoming":[10],"the":[11,44,108,120,131,147,151,154,163],"power":[12,170],"and":[13,29,40,82,113,178],"thermal":[14],"constraints":[15],"that":[16,119,202],"have":[17],"stalled":[18],"single-core":[19],"processor":[20],"scaling.":[21],"By":[22],"integrating":[23],"multi-core":[24,72],"CPUs":[25],"with":[26],"both":[27,38],"discrete":[28,78,121],"integrated":[30,84],"GPUs,":[31],"modern":[32],"systems":[33],"promise":[34],"substantial":[35],"gains":[36],"in":[37,211],"performance":[39],"energy":[41,114,141,148],"efficiency,":[42,142],"yet":[43],"practical":[45],"magnitude":[46],"of":[47,62,107,146],"these":[48],"benefits":[49],"on":[50,214],"consumer":[51],"hardware":[52],"remains":[53],"underexplored.":[54],"This":[55],"study":[56],"presents":[57],"a":[58,63,71,77,90,103,126],"rigorous":[59],"experimental":[60],"comparison":[61],"canonical":[64],"matrix-matrix":[65],"multiplication":[66],"workload":[67,176],"across":[68],"three":[69],"architectures,":[70],"AMD":[73,85],"Ryzen":[74],"7":[75],"CPU,":[76,132],"NVIDIA":[79],"GeForce":[80],"GPU,":[81,88],"Radeon":[86],"Vega":[87],"within":[89],"single,":[91],"widely":[92,215],"available":[93],"laptop.":[94],"Using":[95],"minimally":[96],"intrusive,":[97],"production-grade":[98],"measurement":[99],"tools,":[100],"we":[101],"deliver":[102],"transparent,":[104],"quantitative":[105],"analysis":[106],"real-world":[109],"trade-offs":[110],"between":[111],"speed":[112],"consumption.":[115],"The":[116],"results":[117],"demonstrate":[118],"GPU":[122],"not":[123],"only":[124],"provides":[125],"dramatic":[127],"93-fold":[128],"speedup":[129],"over":[130],"but":[133],"also":[134],"achieves":[135],"more":[136],"than":[137,174],"50":[138],"times":[139],"greater":[140],"consuming":[143],"just":[144],"2%":[145],"required":[149],"by":[150],"CPU":[152],"for":[153,162,183,194,197,207],"same":[155],"computation.":[156],"These":[157],"findings":[158],"provide":[159],"direct":[160],"evidence":[161],"race":[164],"idle":[166,182],"principle:":[167],"peak":[168],"instantaneous":[169],"is":[171,205],"less":[172],"important":[173],"rapid":[175],"completion":[177],"fast":[179],"return":[180],"minimizing":[184],"total":[185],"energy-to-solution.":[186],"Overall,":[187],"this":[188],"work":[189],"establishes":[190],"clear":[191],"empirical":[192],"guidance":[193],"practitioners":[195],"designing":[196],"energy-aware":[198],"high-performance":[199],"computing,":[200],"demonstrating":[201],"architectural":[203],"specialization":[204],"critical":[206],"unlocking":[208],"orders-of-magnitude":[209],"improvements":[210],"computational":[212],"efficiency":[213],"accessible":[216],"platforms.":[217]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":1}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-10-22T00:00:00"}