{"id":"https://openalex.org/W7162400886","doi":"https://doi.org/10.1109/ispass69572.2026.00015","title":"The xPU-athalon: Quantifying the Competition of AI Acceleration","display_name":"The xPU-athalon: Quantifying the Competition of AI Acceleration","publication_year":2026,"publication_date":"2026-04-26","ids":{"openalex":"https://openalex.org/W7162400886","doi":"https://doi.org/10.1109/ispass69572.2026.00015"},"language":null,"primary_location":{"id":"doi:10.1109/ispass69572.2026.00015","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ispass69572.2026.00015","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":"conference-paper","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/A5104179257","display_name":"Alicia Golden","orcid":null},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alicia Golden","raw_affiliation_strings":["Harvard University,Cambridge,MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University,Cambridge,MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028220093","display_name":"Carole-Jean Wu","orcid":"https://orcid.org/0000-0002-9032-7239"},"institutions":[{"id":"https://openalex.org/I2801348669","display_name":"Medieval Academy of America","ror":"https://ror.org/04g3e3p90","country_code":"US","type":"other","lineage":["https://openalex.org/I2801348669"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Carole-Jean Wu","raw_affiliation_strings":["FAIR at Meta,Cambridge,MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAIR at Meta,Cambridge,MA","institution_ids":["https://openalex.org/I2801348669"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043327132","display_name":"Gu-Yeon Wei","orcid":"https://orcid.org/0000-0001-5730-9904"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gu-Yeon Wei","raw_affiliation_strings":["Harvard University,Cambridge,MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University,Cambridge,MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5105205412","display_name":"David Brooks","orcid":null},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Brooks","raw_affiliation_strings":["Harvard University,Cambridge,MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University,Cambridge,MA","institution_ids":["https://openalex.org/I2801851002"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.66832422,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"39","last_page":"50"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.05130000039935112,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.05130000039935112,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.043800000101327896,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11323","display_name":"Gamma-ray bursts and supernovae","score":0.03849999979138374,"subfield":{"id":"https://openalex.org/subfields/3103","display_name":"Astronomy and Astrophysics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.5748999714851379},{"id":"https://openalex.org/keywords/competition","display_name":"Competition (biology)","score":0.4162999987602234},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.3359000086784363},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.2919999957084656}],"concepts":[{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5748999714851379},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4462999999523163},{"id":"https://openalex.org/C91306197","wikidata":"https://www.wikidata.org/wiki/Q45767","display_name":"Competition (biology)","level":2,"score":0.4162999987602234},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.3359000086784363},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.28139999508857727},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2752000093460083},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.2667999863624573},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2378000020980835},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2337000072002411}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ispass69572.2026.00015","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ispass69572.2026.00015","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":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320310174","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305"},{"id":"https://openalex.org/F4320337506","display_name":"Advanced Scientific Computing Research","ror":"https://ror.org/0012c7r22"},{"id":"https://openalex.org/F4320338284","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W3042416028","https://openalex.org/W4281708879","https://openalex.org/W4387321091","https://openalex.org/W4404954357","https://openalex.org/W7133507254"],"related_works":[],"abstract_inverted_index":{"The":[0],"push":[1],"for":[2],"greater":[3],"efficiency":[4,161],"in":[5,65,156],"AI":[6,37],"computation":[7],"has":[8],"given":[9],"rise":[10],"to":[11,158,186],"an":[12],"array":[13],"of":[14,32,36,80,118,125,129,153],"accelerator":[15],"architectures":[16],"that":[17,134],"increasingly":[18],"challenge":[19],"the":[20,40,87,113,126,151,178],"GPU\u2019s":[21],"long-standing":[22],"dominance.":[23],"In":[24],"this":[25,33],"work,":[26],"we":[27,85,164],"provide":[28],"a":[29,101],"quantitative":[30],"view":[31],"evolving":[34],"landscape":[35],"accelerators,":[38],"including":[39],"Cerebras":[41],"CS-3,":[42],"SambaNova":[43],"SN-40,":[44],"Groq,":[45],"Gaudi,":[46],"and":[47,50,57,70,77,97,115,137,147,181],"TPUv5e":[48],"platforms,":[49],"compare":[51],"against":[52],"both":[53,73],"NVIDIA":[54,146],"(A100,":[55],"H100)":[56],"AMD":[58,148],"(MI-300X)":[59],"GPUs.":[60],"We":[61,131],"evaluate":[62],"key":[63],"trade-offs":[64],"latency,":[66],"throughput,":[67],"power":[68,110,144],"consumption,":[69],"energy-efficiency":[71],"across":[72,92,112,167],"(i)":[74],"end-to-end":[75],"workloads":[76],"(ii)":[78],"benchmarks":[79],"individual":[81],"computational":[82],"primitives.":[83],"Notably,":[84],"find":[86,133],"optimal":[88],"hardware":[89],"platform":[90],"varies":[91],"batch":[93],"size,":[94,99],"sequence":[95],"length,":[96],"model":[98],"revealing":[100],"large":[102],"underlying":[103],"optimization":[104],"space.":[105],"Our":[106],"analysis":[107],"includes":[108],"detailed":[109],"measurements":[111],"prefill":[114],"decode":[116],"phases":[117],"LLM":[119],"inference,":[120],"as":[121,123],"well":[122],"quantification":[124],"energy":[127],"cost":[128],"communication.":[130],"additionally":[132],"Cerebras,":[135],"SambaNova,":[136],"Gaudi":[138],"have":[139],"10-60":[140],"%":[141],"higher":[142],"idle":[143],"than":[145],"GPUs,":[149],"emphasizing":[150],"importance":[152],"high":[154],"utilization":[155],"order":[157],"realize":[159],"promised":[160,188],"gains.":[162],"Finally,":[163],"assess":[165],"programmability":[166],"platforms":[168],"based":[169],"on":[170],"our":[171],"experiments":[172],"with":[173],"real":[174],"profiled":[175],"workloads,":[176],"comparing":[177],"compilation":[179],"times":[180],"software":[182],"stack":[183],"maturity":[184],"required":[185],"achieve":[187],"performance.":[189]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-05-27T00:00:00"}