{"id":"https://openalex.org/W2133705126","doi":"https://doi.org/10.1109/asru.2009.5373410","title":"Three-layer optimizations for fast GMM computations on GPU-like parallel processors","display_name":"Three-layer optimizations for fast GMM computations on GPU-like parallel processors","publication_year":2009,"publication_date":"2009-12-01","ids":{"openalex":"https://openalex.org/W2133705126","doi":"https://doi.org/10.1109/asru.2009.5373410","mag":"2133705126"},"language":"en","primary_location":{"id":"doi:10.1109/asru.2009.5373410","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru.2009.5373410","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2009 IEEE Workshop on Automatic Speech Recognition & Understanding","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://escholarship.org/content/qt7z36z8wq/qt7z36z8wq.pdf?t=ptt3x6","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111575885","display_name":"Kshitij Gupta","orcid":null},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Kshitij Gupta","raw_affiliation_strings":["Department of Electrical & Computer Engineering, University of California, Davis, CA, USA","Department of Electrical & Computer Engineering, University of California, Davis, One Shields Avenue, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical & Computer Engineering, University of California, Davis, CA, USA","institution_ids":["https://openalex.org/I84218800"]},{"raw_affiliation_string":"Department of Electrical & Computer Engineering, University of California, Davis, One Shields Avenue, USA","institution_ids":["https://openalex.org/I84218800"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028662746","display_name":"John D. Owens","orcid":"https://orcid.org/0000-0001-6582-8237"},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"John D. Owens","raw_affiliation_strings":["Department of Electrical & Computer Engineering, University of California, Davis, CA, USA","Department of Electrical & Computer Engineering, University of California, Davis, One Shields Avenue, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical & Computer Engineering, University of California, Davis, CA, USA","institution_ids":["https://openalex.org/I84218800"]},{"raw_affiliation_string":"Department of Electrical & Computer Engineering, University of California, Davis, One Shields Avenue, USA","institution_ids":["https://openalex.org/I84218800"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5111575885"],"corresponding_institution_ids":["https://openalex.org/I84218800"],"apc_list":null,"apc_paid":null,"fwci":3.1536,"has_fulltext":false,"cited_by_count":17,"citation_normalized_percentile":{"value":0.92351863,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"146","last_page":"151"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11269","display_name":"Algorithms and Data Compression","score":0.9962999820709229,"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/T11269","display_name":"Algorithms and Data Compression","score":0.9962999820709229,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9886999726295471,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.9822999835014343,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8853930234909058},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7525275945663452},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.7200769186019897},{"id":"https://openalex.org/keywords/locality","display_name":"Locality","score":0.7180025577545166},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6927619576454163},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.5333554744720459},{"id":"https://openalex.org/keywords/locality-of-reference","display_name":"Locality of reference","score":0.5218102931976318},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.43685382604599},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.42134371399879456},{"id":"https://openalex.org/keywords/high-memory","display_name":"High memory","score":0.4198932647705078},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.30080094933509827},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.14360788464546204}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8853930234909058},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7525275945663452},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.7200769186019897},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.7180025577545166},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6927619576454163},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.5333554744720459},{"id":"https://openalex.org/C27602214","wikidata":"https://www.wikidata.org/wiki/Q1868547","display_name":"Locality of reference","level":3,"score":0.5218102931976318},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.43685382604599},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.42134371399879456},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.4198932647705078},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.30080094933509827},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.14360788464546204},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/asru.2009.5373410","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru.2009.5373410","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2009 IEEE Workshop on Automatic Speech Recognition & Understanding","raw_type":"proceedings-article"},{"id":"pmh:qt7z36z8wq","is_oa":true,"landing_page_url":"http://www.escholarship.org/uc/item/7z36z8wq","pdf_url":"https://escholarship.org/content/qt7z36z8wq/qt7z36z8wq.pdf?t=ptt3x6","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,"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":"Gupta, Kshitij; & Owens, John D.(2009). Three-Layer Optimizations for Fast GMM Computations on GPU-like Parallel Processors. Automatic Speech Recognition and Understanding Workshop. doi: 10.1109/ASRU.2009.5373410. UC Davis: Institute for Data Analysis and Visualization. Retrieved from: http://www.escholarship.org/uc/item/7z36z8wq","raw_type":"article"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.362.5756","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.362.5756","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.idav.ucdavis.edu/func/return_pdf?pub_id=976","raw_type":"text"},{"id":"pmh:oai:escholarship.org/ark:/13030/qt7z36z8wq","is_oa":false,"landing_page_url":"https://escholarship.org/uc/item/7z36z8wq","pdf_url":null,"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,"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":"","raw_type":"article"}],"best_oa_location":{"id":"pmh:qt7z36z8wq","is_oa":true,"landing_page_url":"http://www.escholarship.org/uc/item/7z36z8wq","pdf_url":"https://escholarship.org/content/qt7z36z8wq/qt7z36z8wq.pdf?t=ptt3x6","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,"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":"Gupta, Kshitij; & Owens, John D.(2009). Three-Layer Optimizations for Fast GMM Computations on GPU-like Parallel Processors. Automatic Speech Recognition and Understanding Workshop. doi: 10.1109/ASRU.2009.5373410. UC Davis: Institute for Data Analysis and Visualization. Retrieved from: http://www.escholarship.org/uc/item/7z36z8wq","raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320307102","display_name":"Intel Corporation","ror":"https://ror.org/01ek73717"}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2133705126.pdf"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W167388389","https://openalex.org/W323109315","https://openalex.org/W1568041405","https://openalex.org/W1963866266","https://openalex.org/W2101328377","https://openalex.org/W2108157916","https://openalex.org/W2122978030","https://openalex.org/W2138547388","https://openalex.org/W2333081718","https://openalex.org/W6633894228","https://openalex.org/W6702638944"],"related_works":["https://openalex.org/W2031026393","https://openalex.org/W2063611263","https://openalex.org/W3148455537","https://openalex.org/W1905767204","https://openalex.org/W4243474409","https://openalex.org/W2993780376","https://openalex.org/W2607256436","https://openalex.org/W2114927422","https://openalex.org/W2528745529","https://openalex.org/W1966777863"],"abstract_inverted_index":{"In":[0],"this":[1],"paper":[2],"we":[3,46,99],"focus":[4],"on":[5,17,24,42],"optimizing":[6],"compute":[7,104],"and":[8,64,70,105,112],"memory-bandwidth-intensive":[9],"GMM":[10,52,68,126],"computations":[11],"for":[12,86],"low-end,":[13],"small-form-factor":[14],"devices":[15],"running":[16],"GPU-like":[18,43],"parallel":[19,44],"processors.":[20],"With":[21],"special":[22],"emphasis":[23],"tackling":[25],"the":[26,61],"memory":[27,106],"bandwidth":[28,107],"issue":[29],"that":[30],"is":[31],"exacerbated":[32],"by":[33,77],"a":[34,79,90,93],"lack":[35],"of":[36,67,82,89,109],"CPU-like":[37],"caches":[38],"providing":[39],"temporal":[40],"locality":[41,59],"processors,":[45],"propose":[47],"modifications":[48],"to":[49,102,122],"three":[50],"well-known":[51],"computation":[53,127],"reduction":[54],"techniques.":[55,128],"We":[56],"find":[57],"considerable":[58],"at":[60],"frame,":[62],"CI-GMM,":[63],"mixture":[65],"layers":[66],"compute,":[69],"show":[71],"how":[72],"it":[73],"can":[74],"be":[75],"extracted":[76],"following":[78],"chunk-based":[80],"technique":[81],"processing":[83],"multiple":[84],"frames":[85],"every":[87],"load":[88],"GMM.":[91],"On":[92],"1,000-":[94],"word,":[95],"command-and-control,":[96],"continuous-speech":[97],"task,":[98],"are":[100],"able":[101],"achieve":[103],"savings":[108],"over":[110],"60%":[111],"90%":[113],"respectively,":[114],"with":[115],"some":[116],"degradation":[117],"in":[118],"accuracy,":[119],"when":[120],"compared":[121],"existing":[123],"GPU-based":[124],"fast":[125]},"counts_by_year":[{"year":2018,"cited_by_count":1},{"year":2017,"cited_by_count":2},{"year":2016,"cited_by_count":1},{"year":2015,"cited_by_count":3},{"year":2014,"cited_by_count":1},{"year":2013,"cited_by_count":2},{"year":2012,"cited_by_count":5}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}