Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search
Empirical Dynamic Modeling (EDM) is a mathematical framework for modeling and predicting non-linear time series data. Although EDM is increasingly adopted in various research fields, its application to large-scale data has been limited due to its high computational cost. This article presents kEDM,...
Gespeichert in:
Veröffentlicht in: | IEEE access 2023-01, Vol.11, p.1-1 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 1 |
---|---|
container_issue | |
container_start_page | 1 |
container_title | IEEE access |
container_volume | 11 |
creator | Takahashi, Keichi Ichikawa, Kohei Park, Joseph Pao, Gerald M. |
description | Empirical Dynamic Modeling (EDM) is a mathematical framework for modeling and predicting non-linear time series data. Although EDM is increasingly adopted in various research fields, its application to large-scale data has been limited due to its high computational cost. This article presents kEDM, a high-performance implementation of EDM for analyzing large-scale time series datasets. kEDM adopts the Kokkos performance-portable programming model to efficiently run on both CPU and GPU while sharing a single code base. We also conduct hardware-specific optimization of performance-critical kernels. kEDM achieved up to 6.58× speedup in pairwise causal inference of real-world biology datasets compared to an existing EDM implementation. Furthermore, we integrate multiple approximate k-NN search algorithms into EDM to enable the analysis of extremely large datasets that were intractable with conventional EDM based on exhaustive k-NN search. EDM-based time series forecast enhanced with approximate k-NN search demonstrated up to 790× speedup compared to conventional Simplex projection with less than 1% increase in MAPE. |
doi_str_mv | 10.1109/ACCESS.2023.3289836 |
format | Article |
fullrecord | <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_ieee_primary_10164090</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10164090</ieee_id><doaj_id>oai_doaj_org_article_47170377ff1848e5a9abf17cc729ef0d</doaj_id><sourcerecordid>2836054642</sourcerecordid><originalsourceid>FETCH-LOGICAL-c453t-432187538e1dd0be636d6096e5225d47d2cee8fea23980eba8892514429e00293</originalsourceid><addsrcrecordid>eNpNUU1PGzEUXCGQiFJ-QTlY6nmDvz-O0TaUSJQiBdSj5dhvwekmu_VuVPj3OGxU5V38PHozb-wpiq8EzwjB5mZeVYvVakYxZTNGtdFMnhUTSqQpmWDy_KS_LK76foNz6QwJNSl-r7xr3LoBtNh2McV8Q9_fd24bPfrZBmji7gX9i8MrenTJNQ00qGq33X444G4X0LzrUvsWt24A9Kd8eEArcMm_fikuatf0cHU8p8Xz7eKpuivvf_1YVvP70nPBhpIzSrQSTAMJAa9BMhkkNhIEpSJwFagH0DU4yozGsHZaGyoI59QAxtSwabEcdUPrNrZL2Uh6t62L9hNo04t1aYi-AcsVUZgpVddEcw3CGbeuifJeZbEah6z1bdTKL_q7h36wm3afdtm-pflTseCS0zzFximf2r5PUP_fSrA9BGLHQOwhEHsMJLOuR1YEgBMGkRwbzD4Am5GFKw</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2836054642</pqid></control><display><type>article</type><title>Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search</title><source>Free E-Journal (出版社公開部分のみ)</source><source>Directory of Open Access Journals</source><source>IEEE Xplore Open Access Journals</source><creator>Takahashi, Keichi ; Ichikawa, Kohei ; Park, Joseph ; Pao, Gerald M.</creator><creatorcontrib>Takahashi, Keichi ; Ichikawa, Kohei ; Park, Joseph ; Pao, Gerald M.</creatorcontrib><description>Empirical Dynamic Modeling (EDM) is a mathematical framework for modeling and predicting non-linear time series data. Although EDM is increasingly adopted in various research fields, its application to large-scale data has been limited due to its high computational cost. This article presents kEDM, a high-performance implementation of EDM for analyzing large-scale time series datasets. kEDM adopts the Kokkos performance-portable programming model to efficiently run on both CPU and GPU while sharing a single code base. We also conduct hardware-specific optimization of performance-critical kernels. kEDM achieved up to 6.58× speedup in pairwise causal inference of real-world biology datasets compared to an existing EDM implementation. Furthermore, we integrate multiple approximate k-NN search algorithms into EDM to enable the analysis of extremely large datasets that were intractable with conventional EDM based on exhaustive k-NN search. EDM-based time series forecast enhanced with approximate k-NN search demonstrated up to 790× speedup compared to conventional Simplex projection with less than 1% increase in MAPE.</description><identifier>ISSN: 2169-3536</identifier><identifier>EISSN: 2169-3536</identifier><identifier>DOI: 10.1109/ACCESS.2023.3289836</identifier><identifier>CODEN: IAECCG</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Computational modeling ; Cost analysis ; Data analysis ; Datasets ; Dynamic models ; Empirical analysis ; Empirical Dynamic Modeling ; Graphics processing units ; Hardware ; High performance computing ; High-Performance Data Analytics ; Instruction sets ; Optimization ; Parallel processing ; Performance Portability ; Programming ; Search algorithms ; Time series ; Time series analysis</subject><ispartof>IEEE access, 2023-01, Vol.11, p.1-1</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c453t-432187538e1dd0be636d6096e5225d47d2cee8fea23980eba8892514429e00293</citedby><cites>FETCH-LOGICAL-c453t-432187538e1dd0be636d6096e5225d47d2cee8fea23980eba8892514429e00293</cites><orcidid>0000-0003-0094-3984 ; 0000-0001-5411-1409 ; 0000-0002-1607-5694</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10164090$$EHTML$$P50$$Gieee$$Hfree_for_read</linktohtml><link.rule.ids>314,780,784,864,2102,27633,27924,27925,54933</link.rule.ids></links><search><creatorcontrib>Takahashi, Keichi</creatorcontrib><creatorcontrib>Ichikawa, Kohei</creatorcontrib><creatorcontrib>Park, Joseph</creatorcontrib><creatorcontrib>Pao, Gerald M.</creatorcontrib><title>Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search</title><title>IEEE access</title><addtitle>Access</addtitle><description>Empirical Dynamic Modeling (EDM) is a mathematical framework for modeling and predicting non-linear time series data. Although EDM is increasingly adopted in various research fields, its application to large-scale data has been limited due to its high computational cost. This article presents kEDM, a high-performance implementation of EDM for analyzing large-scale time series datasets. kEDM adopts the Kokkos performance-portable programming model to efficiently run on both CPU and GPU while sharing a single code base. We also conduct hardware-specific optimization of performance-critical kernels. kEDM achieved up to 6.58× speedup in pairwise causal inference of real-world biology datasets compared to an existing EDM implementation. Furthermore, we integrate multiple approximate k-NN search algorithms into EDM to enable the analysis of extremely large datasets that were intractable with conventional EDM based on exhaustive k-NN search. EDM-based time series forecast enhanced with approximate k-NN search demonstrated up to 790× speedup compared to conventional Simplex projection with less than 1% increase in MAPE.</description><subject>Computational modeling</subject><subject>Cost analysis</subject><subject>Data analysis</subject><subject>Datasets</subject><subject>Dynamic models</subject><subject>Empirical analysis</subject><subject>Empirical Dynamic Modeling</subject><subject>Graphics processing units</subject><subject>Hardware</subject><subject>High performance computing</subject><subject>High-Performance Data Analytics</subject><subject>Instruction sets</subject><subject>Optimization</subject><subject>Parallel processing</subject><subject>Performance Portability</subject><subject>Programming</subject><subject>Search algorithms</subject><subject>Time series</subject><subject>Time series analysis</subject><issn>2169-3536</issn><issn>2169-3536</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>ESBDL</sourceid><sourceid>RIE</sourceid><sourceid>DOA</sourceid><recordid>eNpNUU1PGzEUXCGQiFJ-QTlY6nmDvz-O0TaUSJQiBdSj5dhvwekmu_VuVPj3OGxU5V38PHozb-wpiq8EzwjB5mZeVYvVakYxZTNGtdFMnhUTSqQpmWDy_KS_LK76foNz6QwJNSl-r7xr3LoBtNh2McV8Q9_fd24bPfrZBmji7gX9i8MrenTJNQ00qGq33X444G4X0LzrUvsWt24A9Kd8eEArcMm_fikuatf0cHU8p8Xz7eKpuivvf_1YVvP70nPBhpIzSrQSTAMJAa9BMhkkNhIEpSJwFagH0DU4yozGsHZaGyoI59QAxtSwabEcdUPrNrZL2Uh6t62L9hNo04t1aYi-AcsVUZgpVddEcw3CGbeuifJeZbEah6z1bdTKL_q7h36wm3afdtm-pflTseCS0zzFximf2r5PUP_fSrA9BGLHQOwhEHsMJLOuR1YEgBMGkRwbzD4Am5GFKw</recordid><startdate>20230101</startdate><enddate>20230101</enddate><creator>Takahashi, Keichi</creator><creator>Ichikawa, Kohei</creator><creator>Park, Joseph</creator><creator>Pao, Gerald M.</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>ESBDL</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0003-0094-3984</orcidid><orcidid>https://orcid.org/0000-0001-5411-1409</orcidid><orcidid>https://orcid.org/0000-0002-1607-5694</orcidid></search><sort><creationdate>20230101</creationdate><title>Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search</title><author>Takahashi, Keichi ; Ichikawa, Kohei ; Park, Joseph ; Pao, Gerald M.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c453t-432187538e1dd0be636d6096e5225d47d2cee8fea23980eba8892514429e00293</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computational modeling</topic><topic>Cost analysis</topic><topic>Data analysis</topic><topic>Datasets</topic><topic>Dynamic models</topic><topic>Empirical analysis</topic><topic>Empirical Dynamic Modeling</topic><topic>Graphics processing units</topic><topic>Hardware</topic><topic>High performance computing</topic><topic>High-Performance Data Analytics</topic><topic>Instruction sets</topic><topic>Optimization</topic><topic>Parallel processing</topic><topic>Performance Portability</topic><topic>Programming</topic><topic>Search algorithms</topic><topic>Time series</topic><topic>Time series analysis</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Takahashi, Keichi</creatorcontrib><creatorcontrib>Ichikawa, Kohei</creatorcontrib><creatorcontrib>Park, Joseph</creatorcontrib><creatorcontrib>Pao, Gerald M.</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005–Present</collection><collection>IEEE Xplore Open Access Journals</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>Directory of Open Access Journals</collection><jtitle>IEEE access</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Takahashi, Keichi</au><au>Ichikawa, Kohei</au><au>Park, Joseph</au><au>Pao, Gerald M.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search</atitle><jtitle>IEEE access</jtitle><stitle>Access</stitle><date>2023-01-01</date><risdate>2023</risdate><volume>11</volume><spage>1</spage><epage>1</epage><pages>1-1</pages><issn>2169-3536</issn><eissn>2169-3536</eissn><coden>IAECCG</coden><abstract>Empirical Dynamic Modeling (EDM) is a mathematical framework for modeling and predicting non-linear time series data. Although EDM is increasingly adopted in various research fields, its application to large-scale data has been limited due to its high computational cost. This article presents kEDM, a high-performance implementation of EDM for analyzing large-scale time series datasets. kEDM adopts the Kokkos performance-portable programming model to efficiently run on both CPU and GPU while sharing a single code base. We also conduct hardware-specific optimization of performance-critical kernels. kEDM achieved up to 6.58× speedup in pairwise causal inference of real-world biology datasets compared to an existing EDM implementation. Furthermore, we integrate multiple approximate k-NN search algorithms into EDM to enable the analysis of extremely large datasets that were intractable with conventional EDM based on exhaustive k-NN search. EDM-based time series forecast enhanced with approximate k-NN search demonstrated up to 790× speedup compared to conventional Simplex projection with less than 1% increase in MAPE.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/ACCESS.2023.3289836</doi><tpages>1</tpages><orcidid>https://orcid.org/0000-0003-0094-3984</orcidid><orcidid>https://orcid.org/0000-0001-5411-1409</orcidid><orcidid>https://orcid.org/0000-0002-1607-5694</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2169-3536 |
ispartof | IEEE access, 2023-01, Vol.11, p.1-1 |
issn | 2169-3536 2169-3536 |
language | eng |
recordid | cdi_ieee_primary_10164090 |
source | Free E-Journal (出版社公開部分のみ); Directory of Open Access Journals; IEEE Xplore Open Access Journals |
subjects | Computational modeling Cost analysis Data analysis Datasets Dynamic models Empirical analysis Empirical Dynamic Modeling Graphics processing units Hardware High performance computing High-Performance Data Analytics Instruction sets Optimization Parallel processing Performance Portability Programming Search algorithms Time series Time series analysis |
title | Scalable Empirical Dynamic Modeling with Parallel Computing and Approximate k-NN Search |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T06%3A00%3A43IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Scalable%20Empirical%20Dynamic%20Modeling%20with%20Parallel%20Computing%20and%20Approximate%20k-NN%20Search&rft.jtitle=IEEE%20access&rft.au=Takahashi,%20Keichi&rft.date=2023-01-01&rft.volume=11&rft.spage=1&rft.epage=1&rft.pages=1-1&rft.issn=2169-3536&rft.eissn=2169-3536&rft.coden=IAECCG&rft_id=info:doi/10.1109/ACCESS.2023.3289836&rft_dat=%3Cproquest_ieee_%3E2836054642%3C/proquest_ieee_%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2836054642&rft_id=info:pmid/&rft_ieee_id=10164090&rft_doaj_id=oai_doaj_org_article_47170377ff1848e5a9abf17cc729ef0d&rfr_iscdi=true |