{"id":"https://openalex.org/W4410583035","doi":"https://doi.org/10.23919/date64628.2025.10992741","title":"DAOP: Data-Aware Offloading and Predictive Pre-Calculation for Efficient MoE Inference","display_name":"DAOP: Data-Aware Offloading and Predictive Pre-Calculation for Efficient MoE Inference","publication_year":2025,"publication_date":"2025-03-31","ids":{"openalex":"https://openalex.org/W4410583035","doi":"https://doi.org/10.23919/date64628.2025.10992741"},"language":"en","primary_location":{"id":"doi:10.23919/date64628.2025.10992741","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date64628.2025.10992741","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Design, Automation & Test in Europe Conference (DATE)","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/A5024638567","display_name":"Yujie Zhang","orcid":"https://orcid.org/0000-0002-1569-0177"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Yujie Zhang","raw_affiliation_strings":["School of Computing, National University of Singapore"],"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101579667","display_name":"Shivam Aggarwal","orcid":"https://orcid.org/0000-0003-1748-9810"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Shivam Aggarwal","raw_affiliation_strings":["School of Computing, National University of Singapore"],"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5049237676","display_name":"Tulika Mitra","orcid":"https://orcid.org/0000-0003-4136-4188"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Tulika Mitra","raw_affiliation_strings":["School of Computing, National University of Singapore"],"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5024638567"],"corresponding_institution_ids":["https://openalex.org/I165932596"],"apc_list":null,"apc_paid":null,"fwci":4.6299,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.94592336,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11361","display_name":"Digital Radiography and Breast Imaging","score":0.8460999727249146,"subfield":{"id":"https://openalex.org/subfields/2740","display_name":"Pulmonary and Respiratory Medicine"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11361","display_name":"Digital Radiography and Breast Imaging","score":0.8460999727249146,"subfield":{"id":"https://openalex.org/subfields/2740","display_name":"Pulmonary and Respiratory Medicine"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.84579998254776,"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/T11719","display_name":"Data Quality and Management","score":0.8259999752044678,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.652059018611908},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6157397031784058},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2429550588130951}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.652059018611908},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6157397031784058},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2429550588130951}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/date64628.2025.10992741","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date64628.2025.10992741","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Design, Automation & Test in Europe Conference (DATE)","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":15,"referenced_works":["https://openalex.org/W2150884987","https://openalex.org/W2946609015","https://openalex.org/W2963339397","https://openalex.org/W2979826702","https://openalex.org/W2998617917","https://openalex.org/W3194676777","https://openalex.org/W4324292875","https://openalex.org/W4389977189","https://openalex.org/W4401211627","https://openalex.org/W4401568723","https://openalex.org/W4402671637","https://openalex.org/W4403006781","https://openalex.org/W4403278766","https://openalex.org/W6682631176","https://openalex.org/W6858205958"],"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":{"Mixture-of-Experts":[0],"(MoE)":[1],"models,":[2],"though":[3],"highly":[4],"effective":[5],"for":[6],"various":[7,100,117],"machine":[8],"learning":[9],"tasks,":[10],"face":[11],"significant":[12],"deployment":[13],"challenges":[14],"on":[15,36,78,87],"memory-constrained":[16],"devices.":[17],"While":[18],"GPUs":[19],"offer":[20],"fast":[21],"inference,":[22],"their":[23],"limited":[24],"memory":[25],"compared":[26],"to":[27,64,89,131],"CPUs":[28,88],"means":[29],"not":[30],"all":[31],"experts":[32,72,86],"can":[33],"be":[34],"stored":[35],"the":[37],"GPU":[38,50,76],"simultaneously,":[39],"necessitating":[40],"frequent,":[41],"costly":[42],"data":[43],"transfers":[44],"from":[45],"CPU":[46,74],"memory,":[47],"often":[48],"negating":[49],"speed":[51],"advantages.":[52],"To":[53],"address":[54],"this,":[55],"we":[56],"present":[57],"DAOP,":[58],"an":[59],"on-device":[60],"MoE":[61],"inference":[62],"engine":[63],"optimize":[65],"parallel":[66],"GPU-CPU":[67],"execution.":[68],"DAOP":[69,121],"dynamically":[70],"allocates":[71],"between":[73],"and":[75,82,126,133],"based":[77],"per-sequence":[79],"activation":[80],"patterns,":[81],"selectively":[83],"pre-calculates":[84],"predicted":[85],"minimize":[90],"transfer":[91],"latency.":[92],"This":[93],"approach":[94],"enables":[95],"efficient":[96],"resource":[97],"utilization":[98],"across":[99,116],"expert":[101,124],"cache":[102],"ratios":[103],"while":[104,138],"maintaining":[105,139],"model":[106],"accuracy":[107],"through":[108],"a":[109],"novel":[110],"graceful":[111],"degradation":[112],"mechanism.":[113],"Comprehensive":[114],"evaluations":[115],"datasets":[118],"show":[119],"that":[120],"outperforms":[122],"traditional":[123],"caching":[125],"prefetching":[127],"methods":[128],"by":[129,136],"up":[130],"8.20x":[132],"offloading":[134],"techniques":[135],"1.35x":[137],"accuracy.":[140]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}