{"id":"https://openalex.org/W4200467895","doi":"https://doi.org/10.1109/hpec49654.2021.9622822","title":"Efficient Neighbor-Sampling-based GNN Training on CPU-FPGA Heterogeneous Platform","display_name":"Efficient Neighbor-Sampling-based GNN Training on CPU-FPGA Heterogeneous Platform","publication_year":2021,"publication_date":"2021-09-20","ids":{"openalex":"https://openalex.org/W4200467895","doi":"https://doi.org/10.1109/hpec49654.2021.9622822"},"language":"en","primary_location":{"id":"doi:10.1109/hpec49654.2021.9622822","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec49654.2021.9622822","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE High Performance Extreme Computing Conference (HPEC)","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/A5101916609","display_name":"Bingyi Zhang","orcid":"https://orcid.org/0000-0002-8115-0814"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Bingyi Zhang","raw_affiliation_strings":["University of Southern California, Los Angeles, USA"],"affiliations":[{"raw_affiliation_string":"University of Southern California, Los Angeles, USA","institution_ids":["https://openalex.org/I1174212"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015036207","display_name":"Sanmukh R. Kuppannagari","orcid":"https://orcid.org/0000-0002-2062-1483"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sanmukh R. Kuppannagari","raw_affiliation_strings":["University of Southern California, Los Angeles, USA"],"affiliations":[{"raw_affiliation_string":"University of Southern California, Los Angeles, USA","institution_ids":["https://openalex.org/I1174212"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042560222","display_name":"Rajgopal Kannan","orcid":"https://orcid.org/0000-0001-8736-3012"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rajgopal Kannan","raw_affiliation_strings":["US Army Research Lab, Los Angeles, USA"],"affiliations":[{"raw_affiliation_string":"US Army Research Lab, Los Angeles, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033166029","display_name":"Viktor K. Prasanna","orcid":"https://orcid.org/0000-0002-1609-8589"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Viktor Prasanna","raw_affiliation_strings":["University of Southern California, Los Angeles, USA"],"affiliations":[{"raw_affiliation_string":"University of Southern California, Los Angeles, USA","institution_ids":["https://openalex.org/I1174212"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5101916609"],"corresponding_institution_ids":["https://openalex.org/I1174212"],"apc_list":null,"apc_paid":null,"fwci":1.6316,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.86941947,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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":0.9803000092506409,"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/T10064","display_name":"Complex Network Analysis Techniques","score":0.975600004196167,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8655861616134644},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8479260206222534},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7197644710540771},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.7037835121154785},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5124202370643616},{"id":"https://openalex.org/keywords/central-processing-unit","display_name":"Central processing unit","score":0.46391287446022034},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.44550102949142456},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.41902539134025574},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.412422239780426},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.2509126663208008},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2100202739238739},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.09363070130348206}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8655861616134644},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8479260206222534},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7197644710540771},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7037835121154785},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5124202370643616},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.46391287446022034},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.44550102949142456},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.41902539134025574},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.412422239780426},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2509126663208008},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2100202739238739},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.09363070130348206},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.0},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpec49654.2021.9622822","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec49654.2021.9622822","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8899999856948853}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2095705004","https://openalex.org/W2807021761","https://openalex.org/W2945827377","https://openalex.org/W2961295589","https://openalex.org/W2963695795","https://openalex.org/W2963757395","https://openalex.org/W2964015378","https://openalex.org/W2964571482","https://openalex.org/W2973152414","https://openalex.org/W2979747168","https://openalex.org/W3000310304","https://openalex.org/W3046757167","https://openalex.org/W3100078588","https://openalex.org/W3100848837","https://openalex.org/W3101553402","https://openalex.org/W3106125969","https://openalex.org/W3133347161","https://openalex.org/W3172512547","https://openalex.org/W4294558607","https://openalex.org/W6674330103","https://openalex.org/W6726873649","https://openalex.org/W6738964360","https://openalex.org/W6748799445","https://openalex.org/W6748856961","https://openalex.org/W6765543928","https://openalex.org/W6776488958"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2076915000","https://openalex.org/W2532502681","https://openalex.org/W3000570965","https://openalex.org/W2518118925","https://openalex.org/W3159273459","https://openalex.org/W3080250604"],"abstract_inverted_index":{"Graph":[0],"neural":[1],"networks":[2],"(GNNs)":[3],"have":[4,27,47],"become":[5],"increasingly":[6],"important":[7],"in":[8,51],"many":[9],"real-world":[10,17],"applications.":[11],"However,":[12],"training":[13,25,35,42,125,162],"GNN":[14,24,41,87,124,148],"on":[15,58,169],"large-scale":[16,59],"graphs":[18],"is":[19],"still":[20],"challenging.":[21],"Many":[22],"sampling-based":[23],"algorithms":[26,67],"been":[28,70],"proposed":[29],"to":[30,80,107,145,159],"facilitate":[31],"the":[32,82,100,103,108,128,152,161,205],"mini-batch":[33],"style":[34],"process.":[36],"The":[37,89,173],"wellknown":[38],"Neighbor-Sampling-base":[39],"(NS)":[40],"algorithms,":[43],"such":[44,66],"as":[45],"GraphSAGE,":[46],"shown":[48],"great":[49],"advantages":[50],"terms":[52],"of":[53,85,102],"accuracy,":[54],"generalization,":[55],"and":[56,96,112,140,155,187],"scalability":[57],"graphs.":[60],"Nevertheless,":[61],"efficient":[62,142,203],"hardware":[63],"acceleration":[64],"for":[65,122],"has":[68],"not":[69],"systematically":[71],"studied.In":[72],"this":[73],"paper,":[74],"we":[75,117],"conduct":[76],"an":[77,134,141,170],"experimental":[78],"study":[79],"understand":[81],"computational":[83],"characteristics":[84],"NS":[86,123],"training.":[88,149],"evaluation":[90,174],"results":[91,175],"show":[92],"that":[93,177],"neighbor":[94,137,153],"sampling":[95,138],"feature":[97],"aggregation":[98],"take":[99],"majority":[101],"execution":[104],"time":[105],"due":[106],"irregular":[109],"memory":[110,114],"accesses":[111],"extensive":[113],"traffic.":[115],"Then,":[116],"propose":[118,151],"a":[119,166],"system":[120,168],"design":[121,180],"by":[126],"exploiting":[127],"CPU-FPGA":[129,179],"heterogeneous":[130],"platform.":[131,194],"We":[132,150,164],"develop":[133],"optimized":[135],"parallel":[136],"implementation":[139],"FPGA":[143,197],"accelerator":[144,198],"enable":[146],"high-throughput":[147],"sharing":[154],"task":[156],"pipelining":[157],"techniques":[158],"improve":[160],"throughput.":[163],"implement":[165],"prototype":[167],"FPGA-equipped":[171],"server.":[172],"demonstrate":[176],"our":[178,196],"achieves":[181],"12\u221221\u00d7":[182],"speedup":[183,191],"than":[184,192,204],"CPU-only":[185],"platform":[186],"0.4":[188],"\u2212":[189],"3.2\u00d7":[190],"CPU-GPU":[193],"Moreover,":[195],"are":[199],"2.3\u00d7":[200],"more":[201],"energy":[202],"GPU":[206],"board.":[207]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}