{"id":"https://openalex.org/W4386709668","doi":"https://doi.org/10.1145/3605573.3605613","title":"Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training","display_name":"Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training","publication_year":2023,"publication_date":"2023-08-07","ids":{"openalex":"https://openalex.org/W4386709668","doi":"https://doi.org/10.1145/3605573.3605613"},"language":"en","primary_location":{"id":"doi:10.1145/3605573.3605613","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3605573.3605613","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd International Conference on Parallel Processing","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/A5055716015","display_name":"Shenggui Li","orcid":"https://orcid.org/0000-0003-2037-2496"},"institutions":[{"id":"https://openalex.org/I168639165","display_name":"Singapore Institute of Technology","ror":"https://ror.org/01v2c2791","country_code":"SG","type":"education","lineage":["https://openalex.org/I168639165"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Shenggui Li","raw_affiliation_strings":["HPC-AI Technology Inc., Singapore"],"raw_orcid":"https://orcid.org/0000-0003-2037-2496","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., Singapore","institution_ids":["https://openalex.org/I168639165"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048207252","display_name":"Hongxin Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongxin Liu","raw_affiliation_strings":["HPC-AI Technology Inc., China"],"raw_orcid":"https://orcid.org/0009-0001-0495-1108","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011091722","display_name":"Zhengda Bian","orcid":"https://orcid.org/0000-0002-1906-1781"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhengda Bian","raw_affiliation_strings":["HPC-AI Technology Inc., China"],"raw_orcid":"https://orcid.org/0000-0002-1906-1781","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069911612","display_name":"Jiarui Fang","orcid":"https://orcid.org/0000-0002-6724-2763"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiarui Fang","raw_affiliation_strings":["HPC-AI Technology Inc., China"],"raw_orcid":"https://orcid.org/0000-0002-6724-2763","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091782857","display_name":"Haichen Huang","orcid":"https://orcid.org/0009-0004-6301-5557"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haichen Huang","raw_affiliation_strings":["HPC-AI Technology Inc., China"],"raw_orcid":"https://orcid.org/0009-0004-6301-5557","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102852473","display_name":"Yuliang Liu","orcid":"https://orcid.org/0009-0003-1878-0199"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuliang Liu","raw_affiliation_strings":["HPC-AI Technology Inc., China"],"raw_orcid":"https://orcid.org/0009-0003-1878-0199","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102708793","display_name":"Boxiang Wang","orcid":"https://orcid.org/0000-0003-3622-6020"},"institutions":[{"id":"https://openalex.org/I168639165","display_name":"Singapore Institute of Technology","ror":"https://ror.org/01v2c2791","country_code":"SG","type":"education","lineage":["https://openalex.org/I168639165"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Boxiang Wang","raw_affiliation_strings":["HPC-AI Technology Inc., Singapore"],"raw_orcid":"https://orcid.org/0000-0003-3622-6020","affiliations":[{"raw_affiliation_string":"HPC-AI Technology Inc., Singapore","institution_ids":["https://openalex.org/I168639165"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100658705","display_name":"Yang You","orcid":"https://orcid.org/0000-0003-2816-4384"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yang You","raw_affiliation_strings":["National University of Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0003-2816-4384","affiliations":[{"raw_affiliation_string":"National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5055716015"],"corresponding_institution_ids":["https://openalex.org/I168639165"],"apc_list":null,"apc_paid":null,"fwci":9.4169,"has_fulltext":false,"cited_by_count":85,"citation_normalized_percentile":{"value":0.98703008,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"766","last_page":"775"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9951000213623047,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9929999709129333,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8444324135780334},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7328181862831116},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5830392241477966},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5163187384605408},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47987809777259827},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4331277906894684},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.428054541349411},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.4240255355834961},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.1509581208229065},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.100791335105896}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8444324135780334},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7328181862831116},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5830392241477966},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5163187384605408},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47987809777259827},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4331277906894684},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.428054541349411},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.4240255355834961},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.1509581208229065},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.100791335105896},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3605573.3605613","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3605573.3605613","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W201315547","https://openalex.org/W1598993252","https://openalex.org/W2012652661","https://openalex.org/W2194775991","https://openalex.org/W2338908902","https://openalex.org/W2612076670","https://openalex.org/W2969388332","https://openalex.org/W2979826702","https://openalex.org/W2991040477","https://openalex.org/W3081168214","https://openalex.org/W3086105743","https://openalex.org/W3204998121","https://openalex.org/W3206832494","https://openalex.org/W4287025700","https://openalex.org/W4287168993"],"related_works":["https://openalex.org/W2317245370","https://openalex.org/W4249323025","https://openalex.org/W198851386","https://openalex.org/W2030310580","https://openalex.org/W947442053","https://openalex.org/W1980160788","https://openalex.org/W2148915962","https://openalex.org/W2283866686","https://openalex.org/W4287182096","https://openalex.org/W27867058"],"abstract_inverted_index":{"The":[0,60],"success":[1],"of":[2,14,20,77],"Transformer":[3],"models":[4],"has":[5,24],"pushed":[6],"the":[7,17,36,41,64,108],"deep":[8,55],"learning":[9,56],"model":[10,78],"scale":[11,73],"to":[12,26,72,80,107,115],"billions":[13],"parameters,":[15],"but":[16],"memory":[18],"limitation":[19],"a":[21,69],"single":[22],"GPU":[23],"led":[25],"an":[27],"urgent":[28],"need":[29],"for":[30,39],"training":[31,79,86,101,118],"on":[32,120],"multi-GPU":[33],"clusters.":[34],"However,":[35],"best":[37],"practice":[38],"choosing":[40],"optimal":[42],"parallel":[43,58,85],"strategy":[44],"is":[45,97],"still":[46],"lacking,":[47],"as":[48,89],"it":[49],"requires":[50],"domain":[51],"expertise":[52],"in":[53],"both":[54],"and":[57,93,96,102],"computing.":[59],"Colossal-AI":[61,111],"system":[62],"addressed":[63],"above":[65],"challenge":[66],"by":[67],"introducing":[68],"unified":[70],"interface":[71],"your":[74],"sequential":[75],"code":[76],"distributed":[81],"environments.":[82],"It":[83],"supports":[84],"methods":[87],"such":[88],"data,":[90],"pipeline,":[91],"tensor,":[92],"sequence":[94],"parallelism":[95],"integrated":[98],"with":[99],"heterogeneous":[100],"zero":[103],"redundancy":[104],"optimizer.":[105],"Compared":[106],"baseline":[109],"system,":[110],"can":[112],"achieve":[113],"up":[114],"2.76":[116],"times":[117],"speedup":[119],"large-scale":[121],"models.":[122]},"counts_by_year":[{"year":2026,"cited_by_count":9},{"year":2025,"cited_by_count":36},{"year":2024,"cited_by_count":26},{"year":2023,"cited_by_count":12},{"year":2022,"cited_by_count":2}],"updated_date":"2026-05-28T09:10:13.091523","created_date":"2025-10-10T00:00:00"}