{"id":"https://openalex.org/W7111329145","doi":"https://doi.org/10.1145/3774899.3775015","title":"GPU Tail Latency Diagnosis for Serverless and HPC Workloads using eBPF","display_name":"GPU Tail Latency Diagnosis for Serverless and HPC Workloads using eBPF","publication_year":2025,"publication_date":"2025-12-09","ids":{"openalex":"https://openalex.org/W7111329145","doi":"https://doi.org/10.1145/3774899.3775015"},"language":null,"primary_location":{"id":"doi:10.1145/3774899.3775015","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3774899.3775015","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th International Workshop on Serverless Computing","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":null,"display_name":"Erfan Darzi","orcid":"https://orcid.org/0000-0002-6003-0119"},"institutions":[{"id":"https://openalex.org/I4210141641","display_name":"IIT@Harvard","ror":"https://ror.org/044hpwe09","country_code":"US","type":"facility","lineage":["https://openalex.org/I30771326","https://openalex.org/I4210141641"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Erfan Darzi","raw_affiliation_strings":["Harvard, Boston, MA, USA"],"raw_orcid":"https://orcid.org/0000-0002-6003-0119","affiliations":[{"raw_affiliation_string":"Harvard, Boston, MA, USA","institution_ids":["https://openalex.org/I4210141641"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Aldo Pareja","orcid":"https://orcid.org/0009-0004-7562-611X"},"institutions":[{"id":"https://openalex.org/I4210110987","display_name":"IIT@MIT","ror":"https://ror.org/01wp8zh54","country_code":"US","type":"facility","lineage":["https://openalex.org/I30771326","https://openalex.org/I4210110987"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aldo Pareja","raw_affiliation_strings":["MIT/IBM, Boston, MA, USA"],"raw_orcid":"https://orcid.org/0009-0004-7562-611X","affiliations":[{"raw_affiliation_string":"MIT/IBM, Boston, MA, USA","institution_ids":["https://openalex.org/I4210110987"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Kaveh Jalilian","orcid":"https://orcid.org/0009-0006-7193-8022"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kaveh Jalilian","raw_affiliation_strings":["Northeastern, Boston, MA, USA"],"raw_orcid":"https://orcid.org/0009-0006-7193-8022","affiliations":[{"raw_affiliation_string":"Northeastern, Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]}]},{"author_position":"last","author":{"id":null,"display_name":"Shreeanant Bharadwaj","orcid":"https://orcid.org/0009-0006-5574-9045"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shreeanant Bharadwaj","raw_affiliation_strings":["Northeastern, Boston, MA, USA"],"raw_orcid":"https://orcid.org/0009-0006-5574-9045","affiliations":[{"raw_affiliation_string":"Northeastern, Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210141641"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.62241645,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"26","last_page":"30"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.32359999418258667,"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":0.32359999418258667,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.27129998803138733,"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/T12127","display_name":"Software System Performance and Reliability","score":0.11089999973773956,"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/pci-express","display_name":"PCI Express","score":0.6901000142097473},{"id":"https://openalex.org/keywords/root-cause","display_name":"Root cause","score":0.6358000040054321},{"id":"https://openalex.org/keywords/debugging","display_name":"Debugging","score":0.6171000003814697},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.522599995136261},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.5019999742507935},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.5001999735832214},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4927999973297119},{"id":"https://openalex.org/keywords/host","display_name":"Host (biology)","score":0.4763999879360199}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8296999931335449},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.6901000142097473},{"id":"https://openalex.org/C84945661","wikidata":"https://www.wikidata.org/wiki/Q7366567","display_name":"Root cause","level":2,"score":0.6358000040054321},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.6171000003814697},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.522599995136261},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.5019999742507935},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.5001999735832214},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4927999973297119},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.48249998688697815},{"id":"https://openalex.org/C126831891","wikidata":"https://www.wikidata.org/wiki/Q221673","display_name":"Host (biology)","level":2,"score":0.4763999879360199},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.45590001344680786},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4016999900341034},{"id":"https://openalex.org/C171078966","wikidata":"https://www.wikidata.org/wiki/Q111029","display_name":"Root (linguistics)","level":2,"score":0.36489999294281006},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.3571999967098236},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.35179999470710754},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.328900009393692},{"id":"https://openalex.org/C32833848","wikidata":"https://www.wikidata.org/wiki/Q4115054","display_name":"Extensibility","level":2,"score":0.30410000681877136},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.2856000065803528},{"id":"https://openalex.org/C200749887","wikidata":"https://www.wikidata.org/wiki/Q1165574","display_name":"System monitoring","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C2778579508","wikidata":"https://www.wikidata.org/wiki/Q722192","display_name":"System call","level":2,"score":0.27000001072883606},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.2563999891281128},{"id":"https://openalex.org/C130795937","wikidata":"https://www.wikidata.org/wiki/Q2561570","display_name":"Remote direct memory access","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3774899.3775015","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3774899.3775015","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th International Workshop on Serverless Computing","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":2,"referenced_works":["https://openalex.org/W4392110350","https://openalex.org/W4394944658"],"related_works":[],"abstract_inverted_index":{"Diagnosing":[0],"GPU":[1,43,86,107],"tail":[2],"latency":[3],"spikes":[4,63],"in":[5,28,72],"serverless":[6],"and":[7,67,99],"multi-tenant":[8,85],"cloud":[9],"environments":[10],"is":[11],"critical":[12],"for":[13,24,52,105],"maintaining":[14],"service-level":[15],"objectives,":[16],"yet":[17],"existing":[18],"monitoring":[19,41],"tools":[20],"lack":[21],"the":[22,89],"granularity":[23],"root":[25,69,92],"cause":[26,70],"analysis":[27,71],"guest-only":[29],"settings.":[30],"We":[31],"introduce":[32],"an":[33],"eBPF-based":[34],"telemetry":[35],"system":[36,54,57,90],"that":[37],"provides":[38],"unified":[39],"host-side":[40],"of":[42],"workloads,":[44,88],"correlating":[45],"eBPF-derived":[46],"host":[47],"metrics":[48],"with":[49,76],"GPU-internal":[50],"events":[51],"holistic":[53],"observability.":[55],"The":[56],"achieves":[58],"81\u201388%":[59],"diagnostic":[60],"accuracy,":[61],"detects":[62],"within":[64],"5":[65],"seconds,":[66,74],"completes":[68],"6\u20138":[73],"operating":[75],"1.21%":[77],"CPU":[78,100],"overhead":[79],"at":[80],"100Hz":[81],"sampling.":[82],"Evaluated":[83],"on":[84],"inference":[87],"identifies":[91],"causes":[93],"including":[94],"NIC":[95],"contention,":[96],"PCIe":[97],"pressure,":[98],"interference,":[101],"enabling":[102],"production":[103],"debugging":[104],"elastic":[106],"platforms":[108],"without":[109],"requiring":[110],"fabric":[111],"privileges.":[112]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-12-10T00:00:00"}
