{"id":"https://openalex.org/W4407217670","doi":"https://doi.org/10.1145/3669940.3707256","title":"vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention","display_name":"vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention","publication_year":2025,"publication_date":"2025-02-06","ids":{"openalex":"https://openalex.org/W4407217670","doi":"https://doi.org/10.1145/3669940.3707256"},"language":"en","primary_location":{"id":"doi:10.1145/3669940.3707256","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707256","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/10.1145/3669940.3707256","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069996639","display_name":"Ramya Prabhu","orcid":"https://orcid.org/0009-0002-7266-5522"},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ramya Prabhu","raw_affiliation_strings":["Microsoft Research, Bangalore, India"],"raw_orcid":"https://orcid.org/0009-0002-7266-5522","affiliations":[{"raw_affiliation_string":"Microsoft Research, Bangalore, India","institution_ids":["https://openalex.org/I4210124949"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084316984","display_name":"A.K. Nayak","orcid":"https://orcid.org/0000-0001-5313-1328"},"institutions":[{"id":"https://openalex.org/I59270414","display_name":"Indian Institute of Science Bangalore","ror":"https://ror.org/04dese585","country_code":"IN","type":"education","lineage":["https://openalex.org/I59270414"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ajay Nayak","raw_affiliation_strings":["Indian Institute of Science, Bangalore, India"],"raw_orcid":"https://orcid.org/0000-0001-5313-1328","affiliations":[{"raw_affiliation_string":"Indian Institute of Science, Bangalore, India","institution_ids":["https://openalex.org/I59270414"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005136390","display_name":"Jayashree Mohan","orcid":"https://orcid.org/0009-0005-5260-3203"},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Jayashree Mohan","raw_affiliation_strings":["Microsoft Research, Bangalore, India"],"raw_orcid":"https://orcid.org/0009-0005-5260-3203","affiliations":[{"raw_affiliation_string":"Microsoft Research, Bangalore, India","institution_ids":["https://openalex.org/I4210124949"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015355530","display_name":"Ramachandran Ramjee","orcid":"https://orcid.org/0000-0003-0007-6040"},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ramachandran Ramjee","raw_affiliation_strings":["Microsoft Research, Bangalore, India"],"raw_orcid":"https://orcid.org/0000-0003-0007-6040","affiliations":[{"raw_affiliation_string":"Microsoft Research, Bangalore, India","institution_ids":["https://openalex.org/I4210124949"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5060296906","display_name":"Ashish Panwar","orcid":"https://orcid.org/0009-0007-0621-4412"},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ashish Panwar","raw_affiliation_strings":["Microsoft Research, Bangalore, India"],"raw_orcid":"https://orcid.org/0009-0007-0621-4412","affiliations":[{"raw_affiliation_string":"Microsoft Research, Bangalore, India","institution_ids":["https://openalex.org/I4210124949"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":38.2318,"has_fulltext":false,"cited_by_count":23,"citation_normalized_percentile":{"value":0.99898322,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1133","last_page":"1150"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.998199999332428,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.998199999332428,"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/T11424","display_name":"Security and Verification in Computing","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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9979000091552734,"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/computer-science","display_name":"Computer science","score":0.6139598488807678},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.4948498606681824},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.15351954102516174},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.12356594204902649}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6139598488807678},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.4948498606681824},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.15351954102516174},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.12356594204902649}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3669940.3707256","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707256","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:10.1145/3669940.3707256","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3669940.3707256","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1583837637","https://openalex.org/W2100926301","https://openalex.org/W2612690371","https://openalex.org/W2904412652","https://openalex.org/W2914304175","https://openalex.org/W2928181291","https://openalex.org/W3166541059","https://openalex.org/W4224308101","https://openalex.org/W4308084044","https://openalex.org/W4387321091","https://openalex.org/W4388858680","https://openalex.org/W4394998892","https://openalex.org/W4401211704"],"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":{"PagedAttention":[0,48],"is":[1],"a":[2,26,65],"popular":[3],"approach":[4],"for":[5],"dynamic":[6],"memory":[7,19,45,54],"allocation":[8,16],"in":[9,36,40],"LLM":[10],"serving":[11],"systems.":[12,38],"It":[13],"enables":[14],"on-demand":[15],"of":[17,56],"GPU":[18],"to":[20,42,62,68],"mitigate":[21],"KV":[22,58],"cache":[23,59],"fragmentation":[24],"-":[25],"phenomenon":[27],"that":[28],"crippled":[29],"the":[30,52,57],"batch":[31],"size":[32],"(and":[33],"consequently":[34],"throughput)":[35],"prior":[37],"However,":[39],"trying":[41],"allocate":[43],"physical":[44],"at":[46],"runtime,":[47],"ends":[49],"up":[50],"changing":[51],"virtual":[53],"layout":[55],"from":[60],"contiguous":[61],"non-contiguous.":[63],"Such":[64],"design":[66],"leads":[67],"non-trivial":[69],"programming":[70],"and":[71],"performance":[72],"overheads.":[73]},"counts_by_year":[{"year":2026,"cited_by_count":11},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}