{"id":"https://openalex.org/W4413757087","doi":"https://doi.org/10.1145/3718958.3754352","title":"ByteScale: Communication-Efficient Scaling of LLM Training with a 2048K Context Length on 16384 GPUs","display_name":"ByteScale: Communication-Efficient Scaling of LLM Training with a 2048K Context Length on 16384 GPUs","publication_year":2025,"publication_date":"2025-08-27","ids":{"openalex":"https://openalex.org/W4413757087","doi":"https://doi.org/10.1145/3718958.3754352"},"language":"en","primary_location":{"id":"doi:10.1145/3718958.3754352","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3718958.3754352","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3718958.3754352","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM SIGCOMM 2025 Conference","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3718958.3754352","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5108863265","display_name":"Houyang Ge","orcid":"https://orcid.org/0009-0001-3367-7486"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Hao Ge","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0001-3367-7486","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017329016","display_name":"Junda Feng","orcid":"https://orcid.org/0000-0003-3664-6615"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junda Feng","raw_affiliation_strings":["ByteDance Inc., Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-3664-6615","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102097114","display_name":"Qi Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi Huang","raw_affiliation_strings":["ByteDance Inc., Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-2143-7494","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039254679","display_name":"Fangcheng Fu","orcid":"https://orcid.org/0000-0003-1658-0380"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fangcheng Fu","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-1658-0380","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059601307","display_name":"Xiaonan Nie","orcid":"https://orcid.org/0000-0001-6766-757X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaonan Nie","raw_affiliation_strings":["ByteDance Inc., San Jose, USA"],"raw_orcid":"https://orcid.org/0000-0001-6766-757X","affiliations":[{"raw_affiliation_string":"ByteDance Inc., San Jose, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036310398","display_name":"Lei Zuo","orcid":"https://orcid.org/0000-0002-7423-1675"},"institutions":[{"id":"https://openalex.org/I58610484","display_name":"Seattle University","ror":"https://ror.org/02jqc0m91","country_code":"US","type":"education","lineage":["https://openalex.org/I58610484"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lei Zuo","raw_affiliation_strings":["ByteDance Inc., Seattle, USA"],"raw_orcid":"https://orcid.org/0000-0002-7423-1675","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, USA","institution_ids":["https://openalex.org/I58610484"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084047386","display_name":"Haibin Lin","orcid":"https://orcid.org/0000-0003-4879-5335"},"institutions":[{"id":"https://openalex.org/I58610484","display_name":"Seattle University","ror":"https://ror.org/02jqc0m91","country_code":"US","type":"education","lineage":["https://openalex.org/I58610484"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haibin Lin","raw_affiliation_strings":["ByteDance Inc., Seattle, USA"],"raw_orcid":"https://orcid.org/0000-0003-4879-5335","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, USA","institution_ids":["https://openalex.org/I58610484"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062357883","display_name":"Bin Cui","orcid":"https://orcid.org/0000-0003-1681-4677"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bin Cui","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1681-4677","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113343221","display_name":"Xin Liu","orcid":"https://orcid.org/0009-0006-5258-3887"},"institutions":[{"id":"https://openalex.org/I58610484","display_name":"Seattle University","ror":"https://ror.org/02jqc0m91","country_code":"US","type":"education","lineage":["https://openalex.org/I58610484"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xin Liu","raw_affiliation_strings":["ByteDance Inc., Seattle, USA"],"raw_orcid":"https://orcid.org/0009-0000-8346-3323","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, USA","institution_ids":["https://openalex.org/I58610484"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5108863265"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":2.2554,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.89641434,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"963","last_page":"978"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9991000294685364,"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":0.9991000294685364,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9926000237464905,"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9911999702453613,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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":0.8188251256942749},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.667514979839325},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6116304993629456},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5467450022697449},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.45119568705558777},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3705536723136902}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8188251256942749},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.667514979839325},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6116304993629456},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5467450022697449},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.45119568705558777},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3705536723136902},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3718958.3754352","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3718958.3754352","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3718958.3754352","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM SIGCOMM 2025 Conference","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2502.21231","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.21231","pdf_url":"https://arxiv.org/pdf/2502.21231","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3718958.3754352","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3718958.3754352","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3718958.3754352","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM SIGCOMM 2025 Conference","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3648843532","display_name":null,"funder_award_id":"U23B2048","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8136954100","display_name":null,"funder_award_id":"2022ZD0116315","funder_id":"https://openalex.org/F4320329860","funder_display_name":"National Science and Technology Major Project"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320324787","display_name":"Peking University","ror":"https://ror.org/02v51f717"},{"id":"https://openalex.org/F4320329860","display_name":"National Science and Technology Major Project","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4413757087.pdf","grobid_xml":"https://content.openalex.org/works/W4413757087.grobid-xml"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W398859631","https://openalex.org/W2914209329","https://openalex.org/W2969388332","https://openalex.org/W3081168214","https://openalex.org/W3086105743","https://openalex.org/W3129831491","https://openalex.org/W3204998121","https://openalex.org/W4224308101","https://openalex.org/W4394745423","https://openalex.org/W4404400729"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W4412817058","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W141820298","https://openalex.org/W2495260952"],"abstract_inverted_index":{"Scaling":[0],"long-context":[1,19],"ability":[2],"is":[3],"essential":[4],"for":[5,65,74],"Large":[6],"Language":[7],"Models":[8],"(LLMs).":[9],"To":[10],"amortize":[11],"the":[12,40,52,63,95],"memory":[13],"consumption":[14],"across":[15],"multiple":[16],"devices":[17,53],"in":[18,70],"training,":[20],"inter-data":[21],"partitioning":[22,28],"(a.k.a.":[23,29],"Data":[24],"Parallelism)":[25,31],"and":[26,45,85,91],"intra-data":[27],"Context":[30],"are":[32],"commonly":[33],"used.":[34],"Current":[35],"training":[36,67,96],"frameworks":[37],"predominantly":[38],"treat":[39],"two":[41],"techniques":[42],"as":[43,54],"orthogonal,":[44],"establish":[46],"static":[47,56,86],"communication":[48,90],"groups":[49],"to":[50],"organize":[51],"a":[55,59],"mesh":[57,87],"(e.g.,":[58],"2D":[60],"mesh).":[61],"However,":[62],"sequences":[64],"LLM":[66],"typically":[68],"vary":[69],"lengths,":[71],"no":[72],"matter":[73],"texts,":[75],"multi-modalities":[76],"or":[77],"reinforcement":[78],"learning.":[79],"The":[80],"mismatch":[81],"between":[82],"data":[83],"heterogeneity":[84],"causes":[88],"redundant":[89],"imbalanced":[92],"computation,":[93],"degrading":[94],"efficiency.":[97]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-25T08:17:42.794288","created_date":"2025-08-28T00:00:00"}