{"id":"https://openalex.org/W7015031870","doi":null,"title":"Scalable and Efficient Systems for Large Deep Learning Models","display_name":"Scalable and Efficient Systems for Large Deep Learning Models","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W7015031870"},"language":"en","primary_location":{"id":"pmh:oai:escholarship.org:ark:/13030/qt97r5r746","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/97r5r746","pdf_url":"https://escholarship.org/content/qt97r5r746/qt97r5r746.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"etd"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://escholarship.org/content/qt97r5r746/qt97r5r746.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5013500109","display_name":"L Zheng","orcid":"https://orcid.org/0000-0002-6611-4612","observed_orcids":["https://orcid.org/0000-0002-6611-4612"]},"institutions":[{"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":true,"raw_author_name":"Zheng, Lianmin","raw_affiliation_strings":["Doctor of Philosophy in Computer Science University of California , Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Doctor of Philosophy in Computer Science University of California , Berkeley","institution_ids":["https://openalex.org/I95457486"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5013500109"],"corresponding_institution_ids":["https://openalex.org/I95457486"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":true,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.7929999828338623,"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":0.7929999828338623,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.06449999660253525,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.032499998807907104,"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/large-language-models","display_name":"large language models","score":0.9800000190734863},{"id":"https://openalex.org/keywords/large-scale-deep-learning","display_name":"large-scale deep learning","score":0.6299999952316284},{"id":"https://openalex.org/keywords/sglang","display_name":"SGLang","score":0.9160000085830688},{"id":"https://openalex.org/keywords/large-language-model-serving","display_name":"large language model serving","score":0.8349999785423279},{"id":"https://openalex.org/keywords/chatbot-arena","display_name":"Chatbot Arena","score":0.8109999895095825},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.9729999899864197},{"id":"https://openalex.org/keywords/alpa","display_name":"ALPA","score":0.4650000035762787},{"id":"https://openalex.org/keywords/distributed-computing-platform","display_name":"distributed computing platform","score":0.421999990940094}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8513000011444092},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6597999930381775},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.6363000273704529},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5945000052452087},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.5609999895095825},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.507099986076355},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4975000023841858},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4745999872684479},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.43970000743865967},{"id":"https://openalex.org/C113843644","wikidata":"https://www.wikidata.org/wiki/Q901882","display_name":"Interface (matter)","level":4,"score":0.4178999960422516},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.4113999903202057},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3612000048160553},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.35519999265670776},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3474999964237213},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.3370000123977661},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.32710000872612},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C164155591","wikidata":"https://www.wikidata.org/wiki/Q2067766","display_name":"Satisfiability modulo theories","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C2779041454","wikidata":"https://www.wikidata.org/wiki/Q870780","display_name":"Chatbot","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.25380000472068787}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:escholarship.org:ark:/13030/qt97r5r746","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/97r5r746","pdf_url":"https://escholarship.org/content/qt97r5r746/qt97r5r746.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"etd"}],"best_oa_location":{"id":"pmh:oai:escholarship.org:ark:/13030/qt97r5r746","doi":null,"is_oa":true,"landing_page_url":"https://escholarship.org/uc/item/97r5r746","pdf_url":"https://escholarship.org/content/qt97r5r746/qt97r5r746.pdf","source":{"id":"https://openalex.org/S4306400115","display_name":"eScholarship (California Digital Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2801248553","host_organization_name":"California Digital Library","host_organization_lineage":["https://openalex.org/I2801248553"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"etd"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7015031870.pdf","grobid_xml":"https://content.openalex.org/works/W7015031870.grobid-xml"},"referenced_works_count":83,"referenced_works":["https://openalex.org/W1598866093","https://openalex.org/W1667652561","https://openalex.org/W1836465849","https://openalex.org/W2186615578","https://openalex.org/W2286929393","https://openalex.org/W2338908902","https://openalex.org/W2384495648","https://openalex.org/W2586850891","https://openalex.org/W2612445135","https://openalex.org/W2786320458","https://openalex.org/W2787998955","https://openalex.org/W2794325560","https://openalex.org/W2884700152","https://openalex.org/W2896457183","https://openalex.org/W2900096133","https://openalex.org/W2915589364","https://openalex.org/W2964137095","https://openalex.org/W2969491671","https://openalex.org/W2973727699","https://openalex.org/W2979044977","https://openalex.org/W3021234081","https://openalex.org/W3030163527","https://openalex.org/W3032631388","https://openalex.org/W3037932933","https://openalex.org/W3040573126","https://openalex.org/W3083410900","https://openalex.org/W3086105743","https://openalex.org/W3088415669","https://openalex.org/W3103621845","https://openalex.org/W3121562065","https://openalex.org/W3130395060","https://openalex.org/W3130421533","https://openalex.org/W3138952127","https://openalex.org/W3163465001","https://openalex.org/W3177813494","https://openalex.org/W3198002980","https://openalex.org/W3201174429","https://openalex.org/W3210871626","https://openalex.org/W3210904070","https://openalex.org/W4200634402","https://openalex.org/W4223908421","https://openalex.org/W4224308101","https://openalex.org/W4281690148","https://openalex.org/W4312983671","https://openalex.org/W4318908031","https://openalex.org/W4322718191","https://openalex.org/W4327810158","https://openalex.org/W4327811957","https://openalex.org/W4360836968","https://openalex.org/W4362707064","https://openalex.org/W4366341216","https://openalex.org/W4367000491","https://openalex.org/W4376167141","https://openalex.org/W4377121468","https://openalex.org/W4378499145","https://openalex.org/W4379260375","https://openalex.org/W4381714872","https://openalex.org/W4384662964","https://openalex.org/W4384918448","https://openalex.org/W4385436532","https://openalex.org/W4385571232","https://openalex.org/W4385571466","https://openalex.org/W4385965642","https://openalex.org/W4386185625","https://openalex.org/W4387430850","https://openalex.org/W4387636003","https://openalex.org/W4388514131","https://openalex.org/W4389523807","https://openalex.org/W4393160302","https://openalex.org/W4395112660","https://openalex.org/W4401042689","https://openalex.org/W4401043184","https://openalex.org/W4402671800","https://openalex.org/W4402727764","https://openalex.org/W4415795271","https://openalex.org/W4415797413","https://openalex.org/W7130841482","https://openalex.org/W7133202489","https://openalex.org/W7133209886","https://openalex.org/W7133214447","https://openalex.org/W7133220558","https://openalex.org/W7133221203","https://openalex.org/W7133233472"],"related_works":["https://openalex.org/W7009534852","https://openalex.org/W4315588221","https://openalex.org/W4401116486","https://openalex.org/W7202283289","https://openalex.org/W7147165704","https://openalex.org/W7015518256","https://openalex.org/W4412868569","https://openalex.org/W4412942455","https://openalex.org/W7008450691","https://openalex.org/W7006461566"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,29,172,180],"machine":[3],"learning":[4,12],"have":[5],"primarily":[6],"been":[7],"driven":[8],"by":[9],"large-scale":[10,109],"deep":[11],"models,":[13,54,95],"particularly":[14],"large":[15,19,94,152,210],"language":[16,153],"models.":[17,211],"The":[18],"scale":[20],"and":[21,40,49,58,78,90,99,121,163,192],"new":[22],"capabilities":[23],"of":[24,52,87,137,209],"these":[25,53],"models":[26,154],"present":[27],"challenges":[28],"designing":[30],"infrastructure":[31],"systems":[32,88],"to":[33,42,65,92,126],"support":[34,93],"their":[35,61],"entire":[36],"lifecycle,":[37],"from":[38],"training":[39],"serving":[41],"evaluation.":[43],"To":[44],"meet":[45],"the":[46,173,206],"high":[47],"computational":[48],"memory":[50],"requirements":[51],"while":[55],"fully":[56],"utilizing":[57],"accurately":[59],"evaluating":[60],"capabilities,":[62],"we":[63,83,103,128,176],"need":[64],"redesign":[66],"many":[67],"system":[68,107,149,204],"components,":[69],"such":[70],"as":[71],"compilers,":[72],"distributed":[73,115],"computing":[74],"platforms,":[75],"programming":[76,161],"systems,":[77],"evaluation":[79,100,174,196],"methods.In":[80],"this":[81],"dissertation,":[82],"introduce":[84,129],"a":[85,106,131,148,158,187,202],"suite":[86],"designed":[89],"built":[91],"covering":[96],"training,":[97,111],"serving,":[98,127],"phases.":[101],"First,":[102],"discuss":[104],"Alpa,":[105],"for":[108,140,150,168,205],"model-parallel":[110],"which":[112,183],"automatically":[113],"generates":[114],"execution":[116],"plans":[117],"integrating":[118],"both":[119,157],"inter-":[120],"intra-operator":[122],"parallelism.":[123],"Moving":[124],"on":[125],"Ansor,":[130],"compiler":[132],"that":[133,155],"produces":[134],"high-performance":[135],"implementations":[136],"tensor":[138],"programs":[139],"various":[141],"hardware":[142],"backends.":[143],"We":[144],"also":[145],"explore":[146],"SGLang,":[147],"deploying":[151],"includes":[156],"flexible":[159],"front-end":[160],"interface":[162],"an":[164,194],"optimized":[165],"back-end":[166],"runtime":[167],"fast":[169],"inference.":[170],"Lastly,":[171],"phase,":[175],"detail":[177],"our":[178],"efforts":[179],"model":[181],"evaluation,":[182],"include":[184],"Chatbot":[185],"Arena,":[186],"crowdsourced":[188],"live":[189],"benchmark":[190],"platform,":[191],"LLM-as-a-Judge,":[193],"automated":[195],"pipeline.":[197],"These":[198],"tools":[199],"collectively":[200],"form":[201],"full-stack":[203],"continuous":[207],"improvement":[208]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}