{"id":"https://openalex.org/W4391986993","doi":"https://doi.org/10.1145/3627535.3638480","title":"POSTER: LLM-PQ:Serving LLM on Heterogeneous Clusters with Phase-Aware Partition and Adaptive Quantization","display_name":"POSTER: LLM-PQ:Serving LLM on Heterogeneous Clusters with Phase-Aware Partition and Adaptive Quantization","publication_year":2024,"publication_date":"2024-02-20","ids":{"openalex":"https://openalex.org/W4391986993","doi":"https://doi.org/10.1145/3627535.3638480"},"language":"en","primary_location":{"id":"doi:10.1145/3627535.3638480","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627535.3638480","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","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/A5087869611","display_name":"Juntao Zhao","orcid":"https://orcid.org/0000-0003-3376-0607"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Juntao Zhao","raw_affiliation_strings":["The University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0003-3376-0607","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633","https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093967737","display_name":"Borui Wan","orcid":"https://orcid.org/0009-0008-5902-1611"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Borui Wan","raw_affiliation_strings":["The University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0009-0008-5902-1611","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633","https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012597518","display_name":"Chuan Wu","orcid":"https://orcid.org/0000-0002-3144-4398"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chuan Wu","raw_affiliation_strings":["The University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-3144-4398","affiliations":[{"raw_affiliation_string":"The University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633","https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013856812","display_name":"Yanghua Peng","orcid":"https://orcid.org/0000-0003-3989-4358"},"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":"Yanghua Peng","raw_affiliation_strings":["ByteDance Inc., Seattle, USA"],"raw_orcid":"https://orcid.org/0000-0003-3989-4358","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Seattle, USA","institution_ids":["https://openalex.org/I58610484"]}]},{"author_position":"last","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"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5087869611"],"corresponding_institution_ids":["https://openalex.org/I177725633","https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":2.2574,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.88928251,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"460","last_page":"462"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9973999857902527,"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":0.9919000267982483,"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/partition","display_name":"Partition (number theory)","score":0.7808876037597656},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7726203203201294},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7041529417037964},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6459364891052246},{"id":"https://openalex.org/keywords/cluster","display_name":"Cluster (spacecraft)","score":0.420645147562027},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3785530924797058},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3662903904914856},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.23815685510635376},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.22170394659042358},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.12775149941444397}],"concepts":[{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.7808876037597656},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7726203203201294},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7041529417037964},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6459364891052246},{"id":"https://openalex.org/C164866538","wikidata":"https://www.wikidata.org/wiki/Q367351","display_name":"Cluster (spacecraft)","level":2,"score":0.420645147562027},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3785530924797058},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3662903904914856},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.23815685510635376},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.22170394659042358},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.12775149941444397},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3627535.3638480","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627535.3638480","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","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":3,"referenced_works":["https://openalex.org/W1583837637","https://openalex.org/W2982041622","https://openalex.org/W4321636575"],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W3161249280","https://openalex.org/W2267059662","https://openalex.org/W2364268683","https://openalex.org/W4388411807","https://openalex.org/W1519906715","https://openalex.org/W2478803962","https://openalex.org/W3169430512","https://openalex.org/W1986418932","https://openalex.org/W2357796999"],"abstract_inverted_index":{"The":[0],"immense":[1],"sizes":[2],"of":[3,38],"Large-scale":[4],"language":[5],"models":[6,22],"(LLMs)":[7],"have":[8],"led":[9],"to":[10,65],"high":[11],"resource":[12],"demand":[13],"and":[14,41,62],"cost":[15],"for":[16],"running":[17],"the":[18,21,48],"models.":[19],"Though":[20],"are":[23],"largely":[24],"served":[25],"using":[26],"uniform":[27],"high-caliber":[28],"GPUs":[29,43],"nowadays,":[30],"utilizing":[31],"a":[32,36,55],"heterogeneous":[33,71],"cluster":[34],"with":[35],"mix":[37],"available":[39],"high-":[40],"low-capacity":[42],"can":[44],"potentially":[45],"substantially":[46],"reduce":[47],"serving":[49,68],"cost.":[50],"This":[51],"paper":[52],"proposes":[53],"LLM-PQ,":[54],"system":[56],"that":[57],"advocates":[58],"adaptive":[59],"model":[60],"quantization":[61],"phase-aware":[63],"partition":[64],"improve":[66],"LLM":[67],"efficiency":[69],"on":[70,76],"GPU":[72],"clusters.":[73],"Extensive":[74],"experiments":[75],"production":[77],"inference":[78],"workloads":[79],"demonstrate":[80],"throughput":[81],"improvement":[82],"in":[83],"inference,":[84],"showing":[85],"great":[86],"advantages":[87],"over":[88],"state-of-the-art":[89],"works.":[90]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":4}],"updated_date":"2026-05-09T13:55:54.758798","created_date":"2025-10-10T00:00:00"}