{"id":"https://openalex.org/W7134252211","doi":"https://doi.org/10.48550/arxiv.2603.06350","title":"MoEless: Efficient MoE LLM Serving via Serverless Computing","display_name":"MoEless: Efficient MoE LLM Serving via Serverless Computing","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7134252211","doi":"https://doi.org/10.48550/arxiv.2603.06350"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2603.06350","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075787386","display_name":"Hanfei Yu","orcid":"https://orcid.org/0000-0001-5790-4981"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yu, Hanfei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106355629","display_name":"Bei Ouyang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ouyang, Bei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128542080","display_name":"Shwai He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Shwai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128547364","display_name":"Ang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Ang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128383607","display_name":"Hao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hao","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5075787386"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.2863999903202057,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.2863999903202057,"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/T14347","display_name":"Big Data and Digital Economy","score":0.1670999974012375,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.12460000067949295,"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/scalability","display_name":"Scalability","score":0.7666000127792358},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7332000136375427},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.6316999793052673},{"id":"https://openalex.org/keywords/expert-system","display_name":"Expert system","score":0.5570999979972839},{"id":"https://openalex.org/keywords/subject-matter-expert","display_name":"Subject-matter expert","score":0.5120999813079834},{"id":"https://openalex.org/keywords/load-balancing","display_name":"Load balancing (electrical power)","score":0.5116999745368958},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5065000057220459},{"id":"https://openalex.org/keywords/inference-engine","display_name":"Inference engine","score":0.43560001254081726}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8133999705314636},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7666000127792358},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7332000136375427},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.6316999793052673},{"id":"https://openalex.org/C58328972","wikidata":"https://www.wikidata.org/wiki/Q184609","display_name":"Expert system","level":2,"score":0.5570999979972839},{"id":"https://openalex.org/C105002631","wikidata":"https://www.wikidata.org/wiki/Q4833645","display_name":"Subject-matter expert","level":3,"score":0.5120999813079834},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.5116999745368958},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5065000057220459},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4526999890804291},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.43560001254081726},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.4334999918937683},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3702000081539154},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.34860000014305115},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.3481999933719635},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3425999879837036},{"id":"https://openalex.org/C2779370713","wikidata":"https://www.wikidata.org/wiki/Q357554","display_name":"Load management","level":2,"score":0.32749998569488525},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.3156000077724457},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.314300000667572},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C548217200","wikidata":"https://www.wikidata.org/wiki/Q251","display_name":"Java","level":2,"score":0.29280000925064087},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.251800000667572}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2603.06350","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2603.06350","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.06350","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2603.06350","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","score":0.6186056137084961,"id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"become":[5,38,71],"a":[6,39,68],"cornerstone":[7],"of":[8,180],"AI,":[9],"driving":[10],"progress":[11],"across":[12,171],"diverse":[13],"domains":[14],"such":[15],"as":[16],"content":[17],"creation,":[18],"search":[19],"and":[20,23,31,85,103,105,131,149,158,168,173,182,193,204],"recommendation":[21],"systems,":[22],"AI-assisted":[24],"workflows.":[25],"To":[26],"alleviate":[27],"extreme":[28],"training":[29],"costs":[30],"advancing":[32],"model":[33],"scales,":[34],"Mixture-of-Experts":[35],"(MoE)":[36],"has":[37],"popular":[40],"backbone":[41],"for":[42],"modern":[43],"LLMs,":[44],"which":[45],"are":[46],"commonly":[47],"served":[48],"in":[49,78,107],"distributed":[50],"deployment":[51],"using":[52],"expert":[53,64,79,89,101,111,128,146,156],"parallelism":[54],"(EP).":[55],"However,":[56],"MoE's":[57],"sparse":[58],"activation":[59],"mechanism":[60],"leads":[61],"to":[62,142,161,210],"severe":[63],"load":[65,90,129,147],"imbalance,":[66],"where":[67],"few":[69],"experts":[70,172],"overloaded":[72],"while":[73],"others":[74],"remain":[75],"idle,":[76],"resulting":[77,106],"stragglers":[80],"that":[81,126,197],"inflate":[82],"inference":[83,133,200,205],"latency":[84,201],"serving":[86,124],"cost.":[87],"Existing":[88],"balancing":[91],"solutions":[92],"assume":[93],"static":[94],"resource":[95],"configurations":[96],"on":[97,178,184],"serverful":[98],"infrastructures,":[99],"limiting":[100],"scalability":[102],"elasticity,":[104],"either":[108],"costly":[109],"real-time":[110],"swapping":[112],"or":[113],"degraded":[114],"generation":[115],"quality.":[116],"We":[117,153],"present":[118],"MoEless,":[119],"the":[120],"first":[121],"serverless":[122,135],"MoE":[123,191],"framework":[125],"mitigates":[127],"imbalance":[130],"accelerates":[132],"via":[134],"experts.":[136],"MoEless":[137,175,198],"employs":[138],"lightweight,":[139],"layer-aware":[140],"predictors":[141],"accurately":[143],"estimate":[144],"incoming":[145],"distributions":[148],"proactively":[150],"identify":[151],"stragglers.":[152],"design":[154],"optimized":[155],"scaling":[157],"placement":[159],"strategies":[160],"maximize":[162],"function":[163],"locality,":[164],"improve":[165],"GPU":[166],"utilization,":[167],"balance":[169],"loads":[170],"GPUs.":[174],"is":[176],"prototyped":[177],"top":[179],"Megatron-LM":[181],"deployed":[183],"an":[185],"eight-GPU":[186],"testbed.":[187],"Experiments":[188],"with":[189],"open-source":[190],"models":[192],"real-world":[194],"workloads":[195],"show":[196],"reduces":[199],"by":[202,207],"43%":[203],"cost":[206],"84%":[208],"compared":[209],"state-of-the-art":[211],"solutions.":[212]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-03-10T00:00:00"}