{"id":"https://openalex.org/W7162403199","doi":"https://doi.org/10.1109/ispass69572.2026.00066","title":"RTG: Exploring the Dynamic Scheduling Space of Real-Time Generative AI workloads on Emerging Heterogeneous Systems","display_name":"RTG: Exploring the Dynamic Scheduling Space of Real-Time Generative AI workloads on Emerging Heterogeneous Systems","publication_year":2026,"publication_date":"2026-04-26","ids":{"openalex":"https://openalex.org/W7162403199","doi":"https://doi.org/10.1109/ispass69572.2026.00066"},"language":null,"primary_location":{"id":"doi:10.1109/ispass69572.2026.00066","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ispass69572.2026.00066","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)","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/A5092072539","display_name":"Rachid Karami","orcid":null},"institutions":[{"id":"https://openalex.org/I204250578","display_name":"University of California, Irvine","ror":"https://ror.org/04gyf1771","country_code":"US","type":"education","lineage":["https://openalex.org/I204250578"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rachid Karami","raw_affiliation_strings":["University of California, Irvine,Irvine,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California, Irvine,Irvine,USA","institution_ids":["https://openalex.org/I204250578"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120035826","display_name":"Rajeev Patwari","orcid":null},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rajeev Patwari","raw_affiliation_strings":["Advanced Micro Devices,San Jose,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Advanced Micro Devices,San Jose,USA","institution_ids":["https://openalex.org/I4210137977"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074768327","display_name":"Hyoukjun Kwon","orcid":"https://orcid.org/0000-0001-9824-1352"},"institutions":[{"id":"https://openalex.org/I204250578","display_name":"University of California, Irvine","ror":"https://ror.org/04gyf1771","country_code":"US","type":"education","lineage":["https://openalex.org/I204250578"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hyoukjun Kwon","raw_affiliation_strings":["University of California, Irvine,Irvine,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California, Irvine,Irvine,USA","institution_ids":["https://openalex.org/I204250578"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086949543","display_name":"Ashish Sirasao","orcid":"https://orcid.org/0009-0004-7828-0871"},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ashish Sirasao","raw_affiliation_strings":["Advanced Micro Devices,San Jose,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Advanced Micro Devices,San Jose,USA","institution_ids":["https://openalex.org/I4210137977"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.83352678,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"641","last_page":"643"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10933","display_name":"Real-Time Systems Scheduling","score":0.6172999739646912,"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/T10933","display_name":"Real-Time Systems Scheduling","score":0.6172999739646912,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.03180000185966492,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.02410000003874302,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.4991999864578247},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.3684999942779541},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.296999990940094},{"id":"https://openalex.org/keywords/dynamic-priority-scheduling","display_name":"Dynamic priority scheduling","score":0.296099990606308}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7106000185012817},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.4991999864578247},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37459999322891235},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3684999942779541},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.30720001459121704},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C107568181","wikidata":"https://www.wikidata.org/wiki/Q5319000","display_name":"Dynamic priority scheduling","level":3,"score":0.296099990606308},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.26899999380111694},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.24629999697208405},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.24390000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ispass69572.2026.00066","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ispass69572.2026.00066","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":6,"referenced_works":["https://openalex.org/W2866634454","https://openalex.org/W3035461736","https://openalex.org/W4387421577","https://openalex.org/W4391623921","https://openalex.org/W4402508150","https://openalex.org/W4402671766"],"related_works":[],"abstract_inverted_index":{"The":[0],"integration":[1],"of":[2],"Large":[3],"Language":[4],"Model":[5],"(LLM)":[6],"agents":[7],"into":[8],"interactive":[9],"applications\u2014driven":[10],"by":[11],"demands":[12],"for":[13,131],"enhanced":[14],"productivity,":[15],"multitasking,":[16],"and":[17,29,57,69,72,78,96,112],"personalization\u2014has":[18],"shifted":[19],"generative":[20],"AI":[21],"toward":[22],"on-device":[23,137],"execution":[24],"to":[25,35,86,109,135],"ensure":[26],"data":[27],"privacy":[28],"availability.":[30],"This":[31],"trend":[32],"gives":[33],"rise":[34],"Real-Time":[36],"Generative":[37],"(RTG)":[38],"workloads,":[39],"in":[40],"which":[41],"compute-intensive":[42],"LLM":[43,64,118],"inference":[44],"runs":[45],"concurrently":[46],"with":[47],"latency-sensitive":[48],"real-time":[49],"tasks":[50],"on":[51],"heterogeneous":[52],"SoCs":[53],"integrating":[54],"CPU,":[55],"GPU,":[56],"NPU":[58],"backends.":[59],"However,":[60],"significant":[61],"variability":[62],"across":[63],"stages":[65],"(prefill":[66],"vs.":[67],"decode)":[68],"input-dependent":[70],"sequence":[71],"context":[73],"lengths":[74],"creates":[75],"a":[76,121],"complex":[77],"dynamic":[79,133],"scheduling":[80,99,134],"space":[81],"that":[82,104],"conventional":[83],"policies":[84],"fail":[85],"handle.":[87],"In":[88],"this":[89],"work,":[90],"we":[91],"define":[92],"realistic":[93],"RTG":[94,138],"scenarios":[95],"evaluate":[97],"six":[98],"policies.":[100],"Our":[101],"results":[102,127],"show":[103],"an":[105],"RTG-aware":[106],"scheduler":[107],"adapts":[108],"hardware-specific":[110],"preferences":[111],"stage-level":[113],"divergence,":[114],"achieving":[115],"near-standalone":[116],"(dedicated-resource)":[117],"performance":[119],"at":[120],"1.83%":[122],"deadline":[123],"violation":[124],"rate.":[125],"These":[126],"highlight":[128],"the":[129],"need":[130],"workload-aware,":[132],"enable":[136],"applications.":[139]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-27T00:00:00"}