{"id":"https://openalex.org/W4403221889","doi":"https://doi.org/10.1145/3640457.3688111","title":"Embedding Optimization for Training Large-scale Deep Learning Recommendation Systems with EMBark","display_name":"Embedding Optimization for Training Large-scale Deep Learning Recommendation Systems with EMBark","publication_year":2024,"publication_date":"2024-10-08","ids":{"openalex":"https://openalex.org/W4403221889","doi":"https://doi.org/10.1145/3640457.3688111"},"language":"en","primary_location":{"id":"doi:10.1145/3640457.3688111","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3640457.3688111","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3640457.3688111","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"18th ACM Conference on Recommender Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3640457.3688111","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022298190","display_name":"Shijie Liu","orcid":"https://orcid.org/0009-0007-5431-5787"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Shijie Liu","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056303339","display_name":"Nan Zheng","orcid":"https://orcid.org/0000-0003-3261-2135"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nan Zheng","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110460157","display_name":"Hui Kang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hui Kang","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107823692","display_name":"Xavier Simmons","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xavier Simmons","raw_affiliation_strings":["Hardware, NVIDIA Corporation, USA"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042280558","display_name":"J. Z. Zhang","orcid":"https://orcid.org/0000-0002-3155-5805"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junjie Zhang","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102004952","display_name":"Matthias Langer","orcid":"https://orcid.org/0000-0003-1776-8000"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matthias Langer","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054452704","display_name":"Wenjing Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenjing Zhu","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031393194","display_name":"Minseok Lee","orcid":"https://orcid.org/0000-0002-8367-1939"},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Minseok Lee","raw_affiliation_strings":["Hardware, NVIDIA Corporation, USA"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020576598","display_name":"Zehuan Wang","orcid":"https://orcid.org/0000-0002-1072-2651"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zehuan Wang","raw_affiliation_strings":["Hardware, NVIDIA Corporation, China"],"affiliations":[{"raw_affiliation_string":"Hardware, NVIDIA Corporation, China","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5022298190"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6237,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.87858072,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"622","last_page":"632"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9998999834060669,"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"}},"topics":[{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9998999834060669,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9860000014305115,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.983299970626831,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7458012104034424},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6403400301933289},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6030792593955994},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.5926734805107117},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5621016025543213},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5253490805625916},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4022008776664734},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.39403632283210754},{"id":"https://openalex.org/keywords/cartography","display_name":"Cartography","score":0.07527884840965271},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.059461385011672974}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7458012104034424},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6403400301933289},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6030792593955994},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.5926734805107117},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5621016025543213},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5253490805625916},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4022008776664734},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.39403632283210754},{"id":"https://openalex.org/C58640448","wikidata":"https://www.wikidata.org/wiki/Q42515","display_name":"Cartography","level":1,"score":0.07527884840965271},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.059461385011672974},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3640457.3688111","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3640457.3688111","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3640457.3688111","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"18th ACM Conference on Recommender Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3640457.3688111","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3640457.3688111","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3640457.3688111","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"18th ACM Conference on Recommender Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403221889.pdf"},"referenced_works_count":41,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W1997136459","https://openalex.org/W2007158453","https://openalex.org/W2210543184","https://openalex.org/W2275236596","https://openalex.org/W2475334473","https://openalex.org/W2512971201","https://openalex.org/W2614794251","https://openalex.org/W2621550233","https://openalex.org/W2899849645","https://openalex.org/W2926767350","https://openalex.org/W2963601856","https://openalex.org/W2972269283","https://openalex.org/W2979719709","https://openalex.org/W2984020950","https://openalex.org/W2998508934","https://openalex.org/W3008591352","https://openalex.org/W3035965352","https://openalex.org/W3036703963","https://openalex.org/W3043023836","https://openalex.org/W3088463703","https://openalex.org/W3093945404","https://openalex.org/W3099878876","https://openalex.org/W3100127252","https://openalex.org/W3101708369","https://openalex.org/W3121395088","https://openalex.org/W3125012172","https://openalex.org/W3138787737","https://openalex.org/W3153687269","https://openalex.org/W3153887672","https://openalex.org/W3155243801","https://openalex.org/W3157531038","https://openalex.org/W3197720002","https://openalex.org/W3200211247","https://openalex.org/W4214658871","https://openalex.org/W4226328099","https://openalex.org/W4296591817","https://openalex.org/W4296591836","https://openalex.org/W4296591840","https://openalex.org/W4318623169","https://openalex.org/W4381329179"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2081900870","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W4375867731","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W2495260952"],"abstract_inverted_index":{"Training":[0],"large-scale":[1],"deep":[2],"learning":[3],"recommendation":[4],"models":[5],"(DLRMs)":[6],"with":[7,108],"embedding":[8,26,53,88],"tables":[9],"stretching":[10],"across":[11],"multiple":[12],"GPUs":[13],"in":[14],"a":[15,18,36,73],"cluster":[16,84],"presents":[17],"unique":[19],"challenge,":[20],"demanding":[21],"the":[22,117],"efficient":[23,103],"scaling":[24],"of":[25,39,139],"operations":[27],"that":[28],"require":[29],"substantial":[30],"memory":[31],"and":[32,55,67,71,112,129,142],"network":[33,38],"bandwidth":[34],"within":[35],"hierarchical":[37,110],"GPUs.":[40],"To":[41],"tackle":[42],"this":[43],"bottleneck,":[44],"we":[45],"introduce":[46],"EMBark\u2014a":[47],"comprehensive":[48],"solution":[49],"aimed":[50],"at":[51,60],"enhancing":[52],"performance":[54],"overall":[56],"DLRM":[57,118,124],"training":[58,119,136],"throughput":[59,137],"scale.":[61],"EMBark":[62,86,131],"empowers":[63],"users":[64],"to":[65,77,96,115,144],"create":[66],"customize":[68],"sharding":[69,75,150],"strategies,":[70],"features":[72],"highly-automated":[74],"planner,":[76],"accelerate":[78],"diverse":[79],"model":[80],"architectures":[81],"on":[82],"different":[83],"configurations.":[85],"groups":[87],"tables,":[89],"considering":[90],"their":[91],"preferred":[92],"communication":[93,98],"compression":[94],"method":[95],"reduce":[97],"overheads":[99],"effectively.":[100],"It":[101],"embraces":[102],"data-parallel":[104],"category":[105],"distribution,":[106],"combined":[107],"topology-aware":[109],"communication,":[111],"pipelining":[113],"support":[114],"maximize":[116],"throughput.":[120],"Across":[121],"four":[122],"representative":[123],"variants":[125],"(DLRM-DCNv2,":[126],"T180,":[127],"T200,":[128],"T510),":[130],"achieves":[132],"an":[133],"average":[134],"end-to-end":[135],"speedup":[138],"1.5":[140],"\u00d7":[141,146],"up":[143],"1.77":[145],"over":[147],"traditional":[148],"table-row-wise":[149],"approaches.":[151]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}