{"id":"https://openalex.org/W4407218922","doi":"https://doi.org/10.1145/3669940.3707285","title":"<scp>Medusa:</scp> Accelerating Serverless LLM Inference with Materialization","display_name":"<scp>Medusa:</scp> Accelerating Serverless LLM Inference with Materialization","publication_year":2025,"publication_date":"2025-02-06","ids":{"openalex":"https://openalex.org/W4407218922","doi":"https://doi.org/10.1145/3669940.3707285"},"language":"en","primary_location":{"id":"doi:10.1145/3669940.3707285","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707285","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3669940.3707285","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5095771121","display_name":"Shaoxun Zeng","orcid":"https://orcid.org/0009-0004-6493-1768"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Shaoxun Zeng","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0004-6493-1768","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101139637","display_name":"Minhui Xie","orcid":"https://orcid.org/0000-0001-6684-8336"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minhui Xie","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-6684-8336","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109082020","display_name":"S.Y. Gao","orcid":"https://orcid.org/0009-0000-4751-0161"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shiwei Gao","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0009-5259-115X","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028549167","display_name":"Youmin Chen","orcid":"https://orcid.org/0000-0003-4171-4299"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youmin Chen","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-4171-4299","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052095173","display_name":"Youyou Lu","orcid":"https://orcid.org/0000-0002-6214-5390"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youyou Lu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-6214-5390","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5095771121"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":11.6204,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.9778993,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"653","last_page":"668"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11614","display_name":"Cloud Data Security Solutions","score":0.9962999820709229,"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/T11614","display_name":"Cloud Data Security Solutions","score":0.9962999820709229,"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/T10270","display_name":"Blockchain Technology Applications and Security","score":0.9932000041007996,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9926999807357788,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/inference","display_name":"Inference","score":0.6862499117851257},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6489095091819763},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.28933778405189514}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6862499117851257},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6489095091819763},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28933778405189514}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3669940.3707285","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707285","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3669940.3707285","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707285","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.41999998688697815}],"awards":[{"id":"https://openalex.org/G2948749109","display_name":null,"funder_award_id":"No. 62332011","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"}],"funders":[{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W2734669649","https://openalex.org/W2734941459","https://openalex.org/W2932772649","https://openalex.org/W2996705685","https://openalex.org/W3012028616","https://openalex.org/W3021115254","https://openalex.org/W3125961627","https://openalex.org/W3130689885","https://openalex.org/W3153413990","https://openalex.org/W3155705005","https://openalex.org/W3204216247","https://openalex.org/W4214610523","https://openalex.org/W4214690606","https://openalex.org/W4220796798","https://openalex.org/W4294904165","https://openalex.org/W4308760184","https://openalex.org/W4384918448","https://openalex.org/W4387321091","https://openalex.org/W4389157038","https://openalex.org/W4392617597","https://openalex.org/W4394871711","https://openalex.org/W4394923129","https://openalex.org/W4394998532","https://openalex.org/W4400377444"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Serverless":[0],"is":[1],"a":[2,51,66],"promising":[3],"paradigm":[4],"to":[5,25,82],"provide":[6],"scalable,":[7],"cost-efficient,":[8],"and":[9,32,58,65],"easy-to-use":[10],"model":[11,19],"inference":[12,20,84],"services.":[13],"However,":[14],"the":[15,26,34,83,88],"cold":[16,92],"start":[17,93],"of":[18,36,91],"functions":[21],"requires":[22],"loading":[23],"models":[24],"devices,":[27],"which":[28,69],"incurs":[29],"high":[30],"latencies":[31],"undermines":[33],"benefits":[35],"serverless":[37],"computing.":[38],"In":[39],"LLMs,":[40],"things":[41],"get":[42],"even":[43],"worse":[44],"since":[45],"two":[46],"extra":[47],"stages":[48,79],"are":[49,80],"introduced:":[50],"KV":[52,63],"cache":[53],"initialization":[54],"stage":[55,68],"that":[56],"profiles":[57],"anticipates":[59],"memory":[60],"reservation":[61],"for":[62,74],"cache,":[64],"capturing":[67],"dynamically":[70],"constructs":[71],"CUDA":[72],"graphs":[73],"different":[75],"batch":[76],"sizes.":[77],"Both":[78],"paramount":[81],"performance,":[85],"but":[86],"become":[87],"main":[89],"culprit":[90],"latency.":[94]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
