{"id":"https://openalex.org/W1970345306","doi":"https://doi.org/10.1145/2312005.2312025","title":"A scalable framework for heterogeneous GPU-based clusters","display_name":"A scalable framework for heterogeneous GPU-based clusters","publication_year":2012,"publication_date":"2012-06-25","ids":{"openalex":"https://openalex.org/W1970345306","doi":"https://doi.org/10.1145/2312005.2312025","mag":"1970345306"},"language":"en","primary_location":{"id":"doi:10.1145/2312005.2312025","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2312005.2312025","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the twenty-fourth annual ACM symposium on Parallelism in algorithms and architectures","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/A5032742296","display_name":"Fengguang Song","orcid":"https://orcid.org/0000-0001-7382-093X"},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"education","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Fengguang Song","raw_affiliation_strings":["University of Tennessee, Knoxville, TN, USA"],"affiliations":[{"raw_affiliation_string":"University of Tennessee, Knoxville, TN, USA","institution_ids":["https://openalex.org/I75027704"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075517045","display_name":"Jack Dongarra","orcid":"https://orcid.org/0000-0003-3247-1782"},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"education","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jack Dongarra","raw_affiliation_strings":["University of Tennessee, Knoxville, TN, USA"],"affiliations":[{"raw_affiliation_string":"University of Tennessee, Knoxville, TN, USA","institution_ids":["https://openalex.org/I75027704"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5032742296"],"corresponding_institution_ids":["https://openalex.org/I75027704"],"apc_list":null,"apc_paid":null,"fwci":9.0554,"has_fulltext":false,"cited_by_count":51,"citation_normalized_percentile":{"value":0.98273524,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"91","last_page":"100"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998999834060669,"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":0.9998999834060669,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9998999834060669,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9994999766349792,"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/computer-science","display_name":"Computer science","score":0.8991197943687439},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7181118130683899},{"id":"https://openalex.org/keywords/thread","display_name":"Thread (computing)","score":0.5582785606384277},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5557093024253845},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5394898056983948},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.4914323091506958},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.48740699887275696},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.44159093499183655},{"id":"https://openalex.org/keywords/symmetric-multiprocessor-system","display_name":"Symmetric multiprocessor system","score":0.42511478066444397},{"id":"https://openalex.org/keywords/runtime-system","display_name":"Runtime system","score":0.4200565814971924},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.18111339211463928},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.17387288808822632}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8991197943687439},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7181118130683899},{"id":"https://openalex.org/C138101251","wikidata":"https://www.wikidata.org/wiki/Q213092","display_name":"Thread (computing)","level":2,"score":0.5582785606384277},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5557093024253845},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5394898056983948},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4914323091506958},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.48740699887275696},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.44159093499183655},{"id":"https://openalex.org/C172430144","wikidata":"https://www.wikidata.org/wiki/Q17111997","display_name":"Symmetric multiprocessor system","level":2,"score":0.42511478066444397},{"id":"https://openalex.org/C2780870223","wikidata":"https://www.wikidata.org/wiki/Q1004415","display_name":"Runtime system","level":2,"score":0.4200565814971924},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.18111339211463928},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.17387288808822632},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/2312005.2312025","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2312005.2312025","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the twenty-fourth annual ACM symposium on Parallelism in algorithms and architectures","raw_type":"proceedings-article"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.466.9268","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.466.9268","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://web.eecs.utk.edu/~song/publication/spaa12.pdf","raw_type":"text"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.9100000262260437,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W94439627","https://openalex.org/W213000350","https://openalex.org/W988231918","https://openalex.org/W1480928214","https://openalex.org/W1483880184","https://openalex.org/W1863162408","https://openalex.org/W1966592804","https://openalex.org/W1969617278","https://openalex.org/W1981534910","https://openalex.org/W1990893960","https://openalex.org/W2047614366","https://openalex.org/W2065102625","https://openalex.org/W2097309756","https://openalex.org/W2121893797","https://openalex.org/W2122619936","https://openalex.org/W2123031400","https://openalex.org/W2139116943","https://openalex.org/W2153375074","https://openalex.org/W2154750565","https://openalex.org/W2157237396","https://openalex.org/W2163950090","https://openalex.org/W2169631286","https://openalex.org/W2170611190","https://openalex.org/W3103849684","https://openalex.org/W4231150350"],"related_works":["https://openalex.org/W2785820621","https://openalex.org/W153259801","https://openalex.org/W2762992606","https://openalex.org/W2072208258","https://openalex.org/W3186309308","https://openalex.org/W3192695480","https://openalex.org/W844148305","https://openalex.org/W2065392434","https://openalex.org/W2102292411","https://openalex.org/W2999423666"],"abstract_inverted_index":{"GPU-based":[0],"heterogeneous":[1,42,47,108,243],"clusters":[2,268],"continue":[3],"to":[4,13,111,122,137,154,175,230,262],"draw":[5],"attention":[6],"from":[7],"vendors":[8],"and":[9,18,37,95,127,131,157,214,222,254,271],"HPC":[10],"users":[11],"due":[12],"their":[14],"high":[15,233,264],"energy":[16],"efficiency":[17],"much":[19,61],"improved":[20],"single-node":[21],"computational":[22],"performance,":[23],"however,":[24],"there":[25],"is":[26,121,260],"little":[27],"parallel":[28],"software":[29],"available":[30],"that":[31,76,98],"can":[32],"utilize":[33],"all":[34,38],"CPU":[35,197,252],"cores":[36,253],"GPUs":[39],"on":[40,113,141,189,241,266],"the":[41,49,65,68,72,80,83,88,139,242],"system":[43,153,168,188,245],"efficiently.":[44],"On":[45],"a":[46,52,55,60,92,100,124,133,148,170,194,200,206,215,232],"cluster,":[48],"performance":[50,66,234,265],"of":[51,67,82,193,196,202,235],"GPU":[53,116,203],"(or":[54,71],"compute":[56,143,163,198,204],"node)":[57],"increases":[58],"in":[59],"faster":[62],"rate":[63],"than":[64],"PCI-Express":[69],"connection":[70],"interconnection":[73],"network)":[74],"such":[75],"communication":[77,102,212,217,223],"eventually":[78],"becomes":[79],"bottleneck":[81],"entire":[84],"system.":[85],"To":[86],"overcome":[87],"bottleneck,":[89],"we":[90,227],"developed":[91],"multi-level":[93],"partitioning":[94],"distribution":[96],"method":[97],"guarantees":[99],"near-optimal":[101],"volume.":[103],"We":[104,145],"have":[105],"also":[106],"extended":[107],"tile":[109],"algorithms":[110],"work":[112],"distributed":[114,149,172],"memory":[115],"clusters.":[117],"Our":[118],"main":[119],"idea":[120],"execute":[123],"serial":[125],"program":[126],"generate":[128],"hybrid-size":[129],"tasks,":[130,156],"follow":[132],"dataflow":[134],"programming":[135],"model":[136],"fire":[138],"tasks":[140,180],"different":[142],"nodes.":[144],"then":[146],"devised":[147],"dynamic":[150,225],"scheduling":[151],"runtime":[152,167,187],"schedule":[155],"transfer":[158],"data":[159,177],"between":[160,179,183],"hybrid":[161],"CPU-GPU":[162],"nodes":[164],"transparently.":[165],"The":[166,186],"employs":[169],"novel":[171],"task-assignment":[173],"protocol":[174],"solve":[176],"dependencies":[178],"without":[181,269],"coordination":[182],"processing":[184],"units.":[185],"each":[190,249],"node":[191],"consists":[192],"number":[195,201],"threads,":[199,205],"task":[207],"generation":[208],"thread,":[209,213],"an":[210],"MPI":[211],"CUDA":[216],"thread.":[218],"By":[219],"overlapping":[220],"computation":[221],"through":[224],"scheduling,":[226],"are":[228],"able":[229,261],"attain":[231,263],"75":[236],"TFlops":[237],"for":[238],"Cholesky":[239],"factorization":[240],"Keeneland":[244],"using":[246],"100":[247],"nodes,":[248],"with":[250],"twelve":[251],"three":[255],"GPUs.":[256],"Moreover,":[257],"our":[258],"framework":[259],"distributed-memory":[267],"GPUs,":[270],"shared-system":[272],"multiGPUs.":[273]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":4},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2019,"cited_by_count":3},{"year":2018,"cited_by_count":1},{"year":2017,"cited_by_count":3},{"year":2016,"cited_by_count":5},{"year":2015,"cited_by_count":5},{"year":2014,"cited_by_count":15},{"year":2013,"cited_by_count":11}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}