{"id":"https://openalex.org/W4401943373","doi":"https://doi.org/10.1109/cloud62652.2024.00063","title":"MediatorDNN: Contention Mitigation for Co-Located DNN Inference Jobs","display_name":"MediatorDNN: Contention Mitigation for Co-Located DNN Inference Jobs","publication_year":2024,"publication_date":"2024-07-07","ids":{"openalex":"https://openalex.org/W4401943373","doi":"https://doi.org/10.1109/cloud62652.2024.00063"},"language":"en","primary_location":{"id":"doi:10.1109/cloud62652.2024.00063","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/cloud62652.2024.00063","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 17th International Conference on Cloud Computing (CLOUD)","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/A5004601038","display_name":"Seyed Morteza Nabavinejad","orcid":"https://orcid.org/0000-0002-5123-6318"},"institutions":[{"id":"https://openalex.org/I107077323","display_name":"Worcester Polytechnic Institute","ror":"https://ror.org/05ejpqr48","country_code":"US","type":"education","lineage":["https://openalex.org/I107077323"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Seyed Morteza Nabavinejad","raw_affiliation_strings":["Worcester Polytechnic Institute,Computer Science Department,Worcester,MA"],"affiliations":[{"raw_affiliation_string":"Worcester Polytechnic Institute,Computer Science Department,Worcester,MA","institution_ids":["https://openalex.org/I107077323"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015719218","display_name":"Sherief Reda","orcid":"https://orcid.org/0000-0001-8232-4516"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sherief Reda","raw_affiliation_strings":["School of Engineering, Brown University,Providence,RI"],"affiliations":[{"raw_affiliation_string":"School of Engineering, Brown University,Providence,RI","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051346938","display_name":"Tian Guo","orcid":"https://orcid.org/0000-0003-0060-2266"},"institutions":[{"id":"https://openalex.org/I107077323","display_name":"Worcester Polytechnic Institute","ror":"https://ror.org/05ejpqr48","country_code":"US","type":"education","lineage":["https://openalex.org/I107077323"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tian Guo","raw_affiliation_strings":["Worcester Polytechnic Institute,Computer Science Department,Worcester,MA"],"affiliations":[{"raw_affiliation_string":"Worcester Polytechnic Institute,Computer Science Department,Worcester,MA","institution_ids":["https://openalex.org/I107077323"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5004601038"],"corresponding_institution_ids":["https://openalex.org/I107077323"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.11831411,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"7694","issue":null,"first_page":"502","last_page":"512"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9969000220298767,"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"}},"topics":[{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9969000220298767,"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/T13553","display_name":"Age of Information Optimization","score":0.9869999885559082,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.984499990940094,"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":0.6998320817947388},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6726124882698059},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.34572434425354004},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.3295116126537323}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6998320817947388},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6726124882698059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34572434425354004},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.3295116126537323}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cloud62652.2024.00063","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/cloud62652.2024.00063","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 17th International Conference on Cloud Computing (CLOUD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4982867890","display_name":null,"funder_award_id":"2105564,2236987","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W1607035479","https://openalex.org/W1836465849","https://openalex.org/W2055328203","https://openalex.org/W2117539524","https://openalex.org/W2183341477","https://openalex.org/W2302255633","https://openalex.org/W2581065617","https://openalex.org/W2603836393","https://openalex.org/W2955060956","https://openalex.org/W2962965915","https://openalex.org/W2963163009","https://openalex.org/W2963821229","https://openalex.org/W2964081807","https://openalex.org/W2982157693","https://openalex.org/W2988410283","https://openalex.org/W3014817752","https://openalex.org/W3016842236","https://openalex.org/W3016939927","https://openalex.org/W3017091196","https://openalex.org/W3026530461","https://openalex.org/W3043038397","https://openalex.org/W3043302551","https://openalex.org/W3043433718","https://openalex.org/W3099561715","https://openalex.org/W3157306683","https://openalex.org/W3214476430","https://openalex.org/W4206643036","https://openalex.org/W4214512541","https://openalex.org/W4214564691","https://openalex.org/W4231332361","https://openalex.org/W4246193833","https://openalex.org/W4289083201","https://openalex.org/W4297775537","https://openalex.org/W4313229743","https://openalex.org/W4318685300","https://openalex.org/W4385623261","https://openalex.org/W4385731844","https://openalex.org/W4385832149","https://openalex.org/W4386707684","https://openalex.org/W6638733343","https://openalex.org/W6676984168","https://openalex.org/W6730956707","https://openalex.org/W6758283263","https://openalex.org/W6766433103","https://openalex.org/W6779103662"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W4402327032","https://openalex.org/W2382290278"],"abstract_inverted_index":{"With":[0],"the":[1,64,121],"increase":[2],"in":[3,50,58],"computing":[4],"power":[5],"of":[6,53,123],"cutting-edge":[7],"hardware":[8,132],"platforms,":[9],"it":[10],"is":[11],"a":[12,20],"common":[13],"practice":[14],"to":[15,32,48,141,148],"run":[16],"multiple":[17],"jobs":[18,54],"on":[19,109,120,130,145],"single":[21],"machine":[22],"for":[23,67],"improved":[24],"resource":[25,34,43,51,78,103,156],"utilization":[26,52,79,104,157],"and":[27,77,97,101,112,154],"throughput.":[28],"However,":[29],"this":[30],"leads":[31],"inevitable":[33],"contention":[35,44,66,76,153],"among":[36],"co-located":[37],"jobs,":[38,70],"impacting":[39],"their":[40,59],"performance.":[41],"The":[42],"can":[45],"worsen":[46],"due":[47],"fluctuations":[49],"caused":[55],"by":[56,139],"variations":[57],"input":[60],"workload.":[61],"To":[62],"tackle":[63],"co-location":[65,122],"DNN":[68,83],"inference":[69,84],"we":[71],"propose":[72],"MediatorDNN,":[73],"which":[74],"considers":[75],"variation":[80],"when":[81],"co-locating":[82],"jobs.":[85,124],"It":[86],"profiles":[87],"each":[88],"DNN,":[89],"monitors":[90],"microarchitectural":[91],"metrics":[92],"such":[93],"as":[94],"memory":[95],"bandwidth":[96],"cache":[98],"access":[99],"pattern,":[100],"high-level":[102],"like":[105],"CPU":[106],"utilization.":[107],"Based":[108],"profiling":[110],"results":[111,126],"leveraging":[113],"Modern":[114],"Portfolio":[115],"Theory":[116],"(MPT),":[117],"MediatorDNN":[118,136],"decides":[119],"Experimental":[125],"with":[127],"various":[128],"DNNs":[129],"two":[131],"platforms":[133],"show":[134],"that":[135],"improves":[137],"throughput":[138],"up":[140],"108%":[142],"(21":[143],"%":[144],"average)":[146],"compared":[147],"an":[149],"approach":[150],"only":[151],"considering":[152],"ignoring":[155],"variation.":[158]},"counts_by_year":[],"updated_date":"2025-12-23T23:11:35.936235","created_date":"2025-10-10T00:00:00"}