{"id":"https://openalex.org/W4318541692","doi":"https://doi.org/10.1145/3575693.3575712","title":"Optimus-CC: Efficient Large NLP Model Training with 3D Parallelism Aware Communication Compression","display_name":"Optimus-CC: Efficient Large NLP Model Training with 3D Parallelism Aware Communication Compression","publication_year":2023,"publication_date":"2023-01-27","ids":{"openalex":"https://openalex.org/W4318541692","doi":"https://doi.org/10.1145/3575693.3575712"},"language":"en","primary_location":{"id":"doi:10.1145/3575693.3575712","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3575693.3575712","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","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/A5101593827","display_name":"Jaeyong Song","orcid":"https://orcid.org/0000-0001-9976-7487"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaeyong Song","raw_affiliation_strings":["Yonsei University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017735447","display_name":"Jinkyu Yim","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinkyu Yim","raw_affiliation_strings":["Seoul National University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University, South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060124284","display_name":"Jaewon Jung","orcid":"https://orcid.org/0000-0002-9081-685X"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaewon Jung","raw_affiliation_strings":["Yonsei University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076710593","display_name":"Hongsun Jang","orcid":"https://orcid.org/0000-0003-4291-6124"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hongsun Jang","raw_affiliation_strings":["Seoul National University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University, South Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100684513","display_name":"Hyungjin Kim","orcid":"https://orcid.org/0000-0002-4834-6882"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyung-Jin Kim","raw_affiliation_strings":["Samsung Electronics, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Electronics, South Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088551237","display_name":"Youngsok Kim","orcid":"https://orcid.org/0000-0002-1015-9969"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youngsok Kim","raw_affiliation_strings":["Yonsei University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yonsei University, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100447633","display_name":"Jinho Lee","orcid":"https://orcid.org/0000-0003-4010-6611"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinho Lee","raw_affiliation_strings":["Seoul National University, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University, South Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":5.4354,"has_fulltext":false,"cited_by_count":34,"citation_normalized_percentile":{"value":0.9638806,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"560","last_page":"573"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9986000061035156,"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/T10028","display_name":"Topic Modeling","score":0.9986000061035156,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9980999827384949,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9972000122070312,"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/computer-science","display_name":"Computer science","score":0.8431273698806763},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.7537007331848145},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.7130908370018005},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.6126537322998047},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.5055525302886963},{"id":"https://openalex.org/keywords/volume","display_name":"Volume (thermodynamics)","score":0.43813902139663696},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.43544313311576843},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.43346381187438965},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.4279385209083557},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42627400159835815},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3786754012107849},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.1367688775062561}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8431273698806763},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.7537007331848145},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.7130908370018005},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.6126537322998047},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.5055525302886963},{"id":"https://openalex.org/C20556612","wikidata":"https://www.wikidata.org/wiki/Q4469374","display_name":"Volume (thermodynamics)","level":2,"score":0.43813902139663696},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.43544313311576843},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.43346381187438965},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.4279385209083557},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42627400159835815},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3786754012107849},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.1367688775062561},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3575693.3575712","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3575693.3575712","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2049668033","https://openalex.org/W2131613942","https://openalex.org/W2775776326","https://openalex.org/W2786602455","https://openalex.org/W2912208425","https://openalex.org/W2951939640","https://openalex.org/W2953212265","https://openalex.org/W2962747323","https://openalex.org/W2963358710","https://openalex.org/W2969388332","https://openalex.org/W2975712713","https://openalex.org/W2981114289","https://openalex.org/W2983101505","https://openalex.org/W3086105743","https://openalex.org/W3092260380","https://openalex.org/W3106929255","https://openalex.org/W3129831491","https://openalex.org/W3132107458","https://openalex.org/W3204998121","https://openalex.org/W3205803342","https://openalex.org/W3206832494","https://openalex.org/W4285826647","https://openalex.org/W4294969081","https://openalex.org/W6893800243","https://openalex.org/W6893809404","https://openalex.org/W6931924280"],"related_works":["https://openalex.org/W2950520577","https://openalex.org/W1501159154","https://openalex.org/W1554644772","https://openalex.org/W2003935582","https://openalex.org/W2494130044","https://openalex.org/W3170887803","https://openalex.org/W74409296","https://openalex.org/W3209384898","https://openalex.org/W4400951174","https://openalex.org/W1595834484"],"abstract_inverted_index":{"In":[0],"training":[1],"of":[2,33],"modern":[3],"large":[4],"natural":[5],"language":[6],"processing":[7],"(NLP)":[8],"models,":[9],"it":[10],"has":[11],"become":[12],"a":[13,30],"common":[14],"practice":[15],"to":[16,22,42,60],"split":[17],"models":[18,65],"using":[19],"3D":[20,67],"parallelism":[21,68,75],"multiple":[23],"GPUs.":[24],"Such":[25],"technique,":[26],"however,":[27,52],"suffers":[28],"from":[29],"high":[31],"overhead":[32,45],"inter-node":[34,49],"communication.":[35],"Compressing":[36],"the":[37,44,48,53,73,81,87],"communication":[38],"is":[39,77],"one":[40],"way":[41],"mitigate":[43],"by":[46],"reducing":[47],"traffic":[50,76],"volume;":[51],"existing":[54,82],"compression":[55,83],"techniques":[56],"have":[57],"critical":[58],"limitations":[59],"be":[61],"applied":[62],"for":[63],"NLP":[64],"with":[66],"in":[69],"that":[70],"1)":[71],"only":[72],"data":[74],"targeted,":[78],"and":[79],"2)":[80],"schemes":[84],"already":[85],"harm":[86],"model":[88],"quality":[89],"too":[90],"much.":[91]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":13},{"year":2023,"cited_by_count":3}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}