{"id":"https://openalex.org/W4416203805","doi":"https://doi.org/10.1145/3712285.3759770","title":"Rethinking Back Transformation in 2-stage Eigenvalue Decomposition on Heterogeneous Architectures","display_name":"Rethinking Back Transformation in 2-stage Eigenvalue Decomposition on Heterogeneous Architectures","publication_year":2025,"publication_date":"2025-11-12","ids":{"openalex":"https://openalex.org/W4416203805","doi":"https://doi.org/10.1145/3712285.3759770"},"language":null,"primary_location":{"id":"doi:10.1145/3712285.3759770","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3712285.3759770","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","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/A5069210933","display_name":"Hansheng Wang","orcid":"https://orcid.org/0009-0006-0035-2323"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hansheng Wang","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0006-0035-2323","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106607687","display_name":"Dajun Huang","orcid":"https://orcid.org/0009-0005-2857-1571"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dajun Huang","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0005-2857-1571","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112284528","display_name":"Guang-Xin Zou","orcid":"https://orcid.org/0009-0001-0550-2662"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Gaoyuan Zou","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0001-0550-2662","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035069795","display_name":"Lu Shi","orcid":"https://orcid.org/0000-0003-2481-0318"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lu Shi","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0003-2481-0318","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051642924","display_name":"Xu Jiang","orcid":"https://orcid.org/0000-0003-2675-2895"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Jiang","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0003-2675-2895","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030763508","display_name":"Xi Wu","orcid":"https://orcid.org/0000-0002-7659-1631"},"institutions":[{"id":"https://openalex.org/I24201400","display_name":"Chengdu University of Information Technology","ror":"https://ror.org/01yxwrh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I24201400"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xi Wu","raw_affiliation_strings":["Chengdu University of Information Technology, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0002-7659-1631","affiliations":[{"raw_affiliation_string":"Chengdu University of Information Technology, Chengdu, China","institution_ids":["https://openalex.org/I24201400"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102079217","display_name":"Hancong Duan","orcid":"https://orcid.org/0000-0002-7721-7422"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hancong Duan","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0002-7721-7422","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5091290657","display_name":"Shaoshuai Zhang","orcid":"https://orcid.org/0000-0002-9525-1659"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaoshuai Zhang","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0002-9525-1659","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.5151,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.83763911,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1830","last_page":"1844"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.31859999895095825,"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":0.31859999895095825,"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.302700012922287,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10792","display_name":"Matrix Theory and Algorithms","score":0.15940000116825104,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/transformation","display_name":"Transformation (genetics)","score":0.7555999755859375},{"id":"https://openalex.org/keywords/eigenvalues-and-eigenvectors","display_name":"Eigenvalues and eigenvectors","score":0.6373000144958496},{"id":"https://openalex.org/keywords/decomposition","display_name":"Decomposition","score":0.544700026512146},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.5095000267028809},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.4918000102043152},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.40880000591278076},{"id":"https://openalex.org/keywords/eigendecomposition-of-a-matrix","display_name":"Eigendecomposition of a matrix","score":0.3824999928474426},{"id":"https://openalex.org/keywords/factor","display_name":"Factor (programming language)","score":0.3691999912261963}],"concepts":[{"id":"https://openalex.org/C204241405","wikidata":"https://www.wikidata.org/wiki/Q461499","display_name":"Transformation (genetics)","level":3,"score":0.7555999755859375},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.6373000144958496},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5910000205039978},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.544700026512146},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.5095000267028809},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5090000033378601},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.4918000102043152},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.45579999685287476},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.40880000591278076},{"id":"https://openalex.org/C169756996","wikidata":"https://www.wikidata.org/wiki/Q194919","display_name":"Eigendecomposition of a matrix","level":3,"score":0.3824999928474426},{"id":"https://openalex.org/C2781039887","wikidata":"https://www.wikidata.org/wiki/Q1391724","display_name":"Factor (programming language)","level":2,"score":0.3691999912261963},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.35910001397132874},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.35659998655319214},{"id":"https://openalex.org/C27665512","wikidata":"https://www.wikidata.org/wiki/Q3042795","display_name":"Benders' decomposition","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.31779998540878296},{"id":"https://openalex.org/C2779791154","wikidata":"https://www.wikidata.org/wiki/Q258040","display_name":"Model transformation","level":3,"score":0.3172000050544739},{"id":"https://openalex.org/C183222429","wikidata":"https://www.wikidata.org/wiki/Q42278","display_name":"Magma","level":3,"score":0.3109000027179718},{"id":"https://openalex.org/C54940322","wikidata":"https://www.wikidata.org/wiki/Q3997740","display_name":"Orthogonal transformation","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.2937999963760376},{"id":"https://openalex.org/C2778258933","wikidata":"https://www.wikidata.org/wiki/Q16918986","display_name":"Decomposition method (queueing theory)","level":2,"score":0.2854999899864197},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.2827000021934509},{"id":"https://openalex.org/C2778361913","wikidata":"https://www.wikidata.org/wiki/Q7248437","display_name":"Program transformation","level":2,"score":0.27230000495910645},{"id":"https://openalex.org/C165443888","wikidata":"https://www.wikidata.org/wiki/Q1482183","display_name":"Transformation matrix","level":3,"score":0.26409998536109924},{"id":"https://openalex.org/C197115733","wikidata":"https://www.wikidata.org/wiki/Q1003136","display_name":"Forcing (mathematics)","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3712285.3759770","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3712285.3759770","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2568803418","display_name":null,"funder_award_id":"U2333211","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W76807688","https://openalex.org/W995198334","https://openalex.org/W2039778940","https://openalex.org/W2053168940","https://openalex.org/W2056805806","https://openalex.org/W2072396164","https://openalex.org/W2073260424","https://openalex.org/W2078464328","https://openalex.org/W2083154122","https://openalex.org/W2094882084","https://openalex.org/W2104110430","https://openalex.org/W2108417208","https://openalex.org/W2119522963","https://openalex.org/W2122860281","https://openalex.org/W2128728535","https://openalex.org/W2148678374","https://openalex.org/W2153757756","https://openalex.org/W2164895445","https://openalex.org/W2294641594","https://openalex.org/W2397480569","https://openalex.org/W2887881931","https://openalex.org/W2951445059","https://openalex.org/W3002341896","https://openalex.org/W3007543653","https://openalex.org/W3008771456","https://openalex.org/W3039141850","https://openalex.org/W3130554079","https://openalex.org/W3154152825","https://openalex.org/W4324292875","https://openalex.org/W4408049835"],"related_works":[],"abstract_inverted_index":{"The":[0],"2-stage":[1],"eigenvalue":[2],"decomposition":[3],"(EVD)":[4],"method":[5,9],"outperforms":[6],"conventional":[7],"1-stage":[8],"on":[10],"GPUs":[11],"and":[12,56,126,133],"heterogeneous":[13],"architectures,":[14],"especially":[15],"when":[16,26],"eigenvectors":[17],"are":[18],"not":[19],"required.":[20],"However,":[21],"its":[22],"performance":[23,92],"advantage":[24],"diminishes":[25],"performing":[27],"back":[28,51,60,87],"transformation":[29,52,61,88],"to":[30,67,72],"obtain":[31],"eigenvectors.":[32],"To":[33],"address":[34],"this,":[35],"we":[36],"propose":[37],"two":[38],"key":[39],"solutions:":[40],"1)":[41],"replacing":[42],"BLAS3":[43],"operations":[44,47],"with":[45,112],"BLAS2":[46],"during":[48],"the":[49,59,77,85,95,114,121],"bulge-chasing":[50],"for":[53],"better":[54],"performance,":[55],"2)":[57],"reordering":[58],"workflow":[62],"from":[63],"a":[64,68],"backward":[65],"pattern":[66,71],"new":[69],"parallelism-driven":[70],"hide":[73],"divide-and-conquer":[74],"latency,":[75],"at":[76],"cost":[78],"of":[79,104,131],"one":[80],"additional":[81],"GEMM":[82],"computation.":[83],"Experimentally,":[84],"proposed":[86,115],"algorithm":[89],"demonstrates":[90],"significant":[91],"improvements,":[93],"outperforming":[94],"SOTA":[96,122],"implementation":[97],"in":[98,124],"MAGMA":[99,125],"by":[100,128],"an":[101],"average":[102,129],"factor":[103],"3.58x.":[105],"For":[106],"complete":[107],"FP64":[108],"precision":[109],"symmetric":[110],"EVD":[111],"eigenvectors,":[113],"algorithm,":[116],"incorporating":[117],"both":[118],"solutions,":[119],"surpasses":[120],"implementations":[123],"cuSOLVER":[127],"factors":[130],"2.62x":[132],"2.21x,":[134],"respectively.":[135]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-12T00:00:00"}
