{"id":"https://openalex.org/W7155497146","doi":"https://doi.org/10.1145/3767295.3769353","title":"SkyWalker: A Locality-Aware Cross-Region Load Balancer for LLM Inference","display_name":"SkyWalker: A Locality-Aware Cross-Region Load Balancer for LLM Inference","publication_year":2026,"publication_date":"2026-04-24","ids":{"openalex":"https://openalex.org/W7155497146","doi":"https://doi.org/10.1145/3767295.3769353"},"language":null,"primary_location":{"id":"doi:10.1145/3767295.3769353","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769353","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 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3767295.3769353","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5091205267","display_name":"Tian Xia","orcid":"https://orcid.org/0000-0003-0123-1305"},"institutions":[{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tian Xia","raw_affiliation_strings":["UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0009-0008-0895-4585","affiliations":[{"raw_affiliation_string":"UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122978591","display_name":"Ziming Mao","orcid":null},"institutions":[{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziming Mao","raw_affiliation_strings":["UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0000-0001-7738-2498","affiliations":[{"raw_affiliation_string":"UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5105100627","display_name":"Jamison Kerney","orcid":null},"institutions":[{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jamison Kerney","raw_affiliation_strings":["UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0009-0007-8647-2680","affiliations":[{"raw_affiliation_string":"UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077155634","display_name":"Ethan J. Jackson","orcid":null},"institutions":[{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ethan J. Jackson","raw_affiliation_strings":["UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0009-0004-7770-0571","affiliations":[{"raw_affiliation_string":"UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100782890","display_name":"Zhifei Li","orcid":"https://orcid.org/0000-0001-5045-0569"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhifei Li","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0003-1488-4871","affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053453103","display_name":"Jiarong Xing","orcid":"https://orcid.org/0009-0006-6163-0569"},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiarong Xing","raw_affiliation_strings":["Rice University AND UC Berkeley, Houston, USA"],"raw_orcid":"https://orcid.org/0009-0006-6163-0569","affiliations":[{"raw_affiliation_string":"Rice University AND UC Berkeley, Houston, USA","institution_ids":["https://openalex.org/I74775410"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122948951","display_name":"Scott Shenker","orcid":null},"institutions":[{"id":"https://openalex.org/I1297971548","display_name":"International Computer Science Institute","ror":"https://ror.org/01ewh7m12","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I1297971548"]},{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Scott Shenker","raw_affiliation_strings":["ICSI AND UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0000-0002-1357-7533","affiliations":[{"raw_affiliation_string":"ICSI AND UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I1297971548"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5041920173","display_name":"Ion Stoica","orcid":"https://orcid.org/0000-0002-5373-0088"},"institutions":[{"id":"https://openalex.org/I134446601","display_name":"Berkeley College","ror":"https://ror.org/02xewxa75","country_code":"US","type":"education","lineage":["https://openalex.org/I134446601"]},{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ion Stoica","raw_affiliation_strings":["UC Berkeley, Berkeley, USA"],"raw_orcid":"https://orcid.org/0000-0002-5373-0088","affiliations":[{"raw_affiliation_string":"UC Berkeley, Berkeley, USA","institution_ids":["https://openalex.org/I134446601","https://openalex.org/I95457486"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1349","last_page":"1364"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.06870000064373016,"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"}},"topics":[{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.06870000064373016,"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"}},{"id":"https://openalex.org/T12205","display_name":"Time Series Analysis and Forecasting","score":0.03849999979138374,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.029100000858306885,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.3384999930858612},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.29179999232292175},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.273499995470047},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.25459998846054077}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5702000260353088},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3756999969482422},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3384999930858612},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3050999939441681},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.29179999232292175},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2685999870300293},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.25609999895095825},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.25459998846054077},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2524000108242035},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3767295.3769353","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769353","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 21st European Conference on Computer Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3767295.3769353","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769353","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 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/13","score":0.4053497016429901,"display_name":"Climate action"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W2808103092","https://openalex.org/W3127327396","https://openalex.org/W4380874786","https://openalex.org/W4397009504","https://openalex.org/W4409248734"],"related_works":[],"abstract_inverted_index":{"Serving":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"efficiently":[5],"in":[6,23,91],"multi-region":[7,58],"setups":[8],"remains":[9],"a":[10,57,113,119],"challenge.":[11],"Due":[12],"to":[13,42,79,143],"cost":[14,104,151],"and":[15,47,100,118,138],"GPU":[16],"availability":[17],"concerns,":[18],"providers":[19,78],"typically":[20],"deploy":[21],"LLMs":[22],"multiple":[24],"regions":[25],"using":[26],"instance":[27],"with":[28,112],"long-term":[29],"commitments,":[30],"like":[31],"reserved":[32],"instances":[33,81],"or":[34],"on-premise":[35],"clusters,":[36],"which":[37],"are":[38],"often":[39],"underutilized":[40],"due":[41],"their":[43],"region-local":[44],"traffic":[45,49,71,116],"handling":[46],"diurnal":[48,67],"variance.":[50],"In":[51],"this":[52,111],"paper,":[53],"we":[54],"introduce":[55],"SkyWalker,":[56],"load":[59,101,123,145],"balancer":[60],"for":[61],"LLM":[62],"inference":[63],"that":[64,132],"aggregates":[65],"regional":[66],"patterns":[68],"through":[69],"cross-region":[70,115],"handling.":[72],"By":[73],"doing":[74],"so,":[75],"SkyWalker":[76,96,109],"enables":[77],"reserve":[80],"based":[82,122],"on":[83,128],"expected":[84],"global":[85],"demand,":[86],"rather":[87],"than":[88],"peak":[89],"demand":[90],"each":[92],"individual":[93],"region.":[94],"Meanwhile,":[95],"preserves":[97],"KV-Cache":[98],"locality":[99],"balancing,":[102],"ensuring":[103],"efficiency":[105],"without":[106],"sacrificing":[107],"performance.":[108],"achieves":[110,134],"cache-aware":[114],"handler":[117],"selective":[120],"pushing":[121],"balancing":[124],"mechanism.":[125],"Our":[126],"evaluation":[127],"real-world":[129],"workloads":[130],"shows":[131],"it":[133],"1.12\u20132.06\u00d7":[135],"higher":[136],"throughput":[137],"1.74\u20136.30\u00d7":[139],"lower":[140],"latency":[141],"compared":[142],"existing":[144],"balancers,":[146],"while":[147],"reducing":[148],"total":[149],"serving":[150],"by":[152],"25%.":[153]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-04-25T00:00:00"}
