{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,3,2]],"date-time":"2025-03-02T05:32:32Z","timestamp":1740893552402,"version":"3.38.0"},"reference-count":33,"publisher":"SAGE Publications","issue":"3","license":[{"start":{"date-parts":[[2020,8,26]],"date-time":"2020-08-26T00:00:00Z","timestamp":1598400000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/501100004834","name":"Universitat Jaume I","doi-asserted-by":"publisher","award":["POSDOC-A\/2017\/11"],"award-info":[{"award-number":["POSDOC-A\/2017\/11"]}],"id":[{"id":"10.13039\/501100004834","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003359","name":"Generalitat Valenciana","doi-asserted-by":"publisher","award":["CDEIGENT\/2018\/014"],"award-info":[{"award-number":["CDEIGENT\/2018\/014"]}],"id":[{"id":"10.13039\/501100003359","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003329","name":"Ministerio de Econom\u00eda y Competitividad","doi-asserted-by":"publisher","award":["TIN2017-82972-R"],"award-info":[{"award-number":["TIN2017-82972-R"]}],"id":[{"id":"10.13039\/501100003329","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of High Performance Computing Applications"],"published-print":{"date-parts":[[2021,5]]},"abstract":"<jats:p> Modeling the performance and energy consumption of the sparse matrix-vector product (SpMV) is essential to perform off-line analysis and, for example, choose a target computer architecture that delivers the best performance-energy consumption ratio. However, this task is especially complex given the memory-bounded nature and irregular memory accesses of the SpMV, mainly dictated by the input sparse matrix. In this paper, we propose a Machine Learning (ML)-driven approach that leverages Convolutional Neural Networks (CNNs) to provide accurate estimations of the performance and energy consumption of the SpMV kernel. The proposed CNN-based models use a blockwise approach to make the CNN architecture independent of the matrix size. These models are trained to estimate execution time as well as total, package, and DRAM energy consumption at different processor frequencies. The experimental results reveal that the overall relative error ranges between 0.5% and 14%, while at matrix level is not superior to 10%. To demonstrate the applicability and accuracy of the SpMV CNN-based models, this study is complemented with an ad-hoc time-energy model for the PageRank algorithm, a popular algorithm for web information retrieval used by search engines, which internally realizes the SpMV kernel. <\/jats:p>","DOI":"10.1177\/1094342020953196","type":"journal-article","created":{"date-parts":[[2020,8,26]],"date-time":"2020-08-26T11:55:12Z","timestamp":1598442912000},"page":"268-281","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":3,"title":["Convolutional neural nets for estimating the run time and energy consumption of the sparse matrix-vector product"],"prefix":"10.1177","volume":"35","author":[{"given":"Maria","family":"Barreda","sequence":"first","affiliation":[{"name":"Departament d\u2019Enginyeria i Ci\u00e8ncia dels Computadors, Universitat Jaume I de Castell\u00f3, Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9466-3398","authenticated-orcid":false,"given":"Manuel F","family":"Dolz","sequence":"additional","affiliation":[{"name":"Departament d\u2019Enginyeria i Ci\u00e8ncia dels Computadors, Universitat Jaume I de Castell\u00f3, Spain"}]},{"given":"M Asunci\u00f3n","family":"Casta\u00f1o","sequence":"additional","affiliation":[{"name":"Departament d\u2019Enginyeria i Ci\u00e8ncia dels Computadors, Universitat Jaume I de Castell\u00f3, Spain"}]}],"member":"179","published-online":{"date-parts":[[2020,8,26]]},"reference":[{"key":"bibr1-1094342020953196","unstructured":"Abadi M, Agarwal A, Barham P, et al. (2015) TensorFlow: large-scale machine learning on heterogeneous systems. Available at: https:\/\/www.tensorflow.org\/. Software available from tensorflow.org (accessed July 2020)."},{"key":"bibr2-1094342020953196","doi-asserted-by":"crossref","unstructured":"Barreda M, Dolz MF, Casta\u00f1o MA (2020a) Block-wise sparse matrix-vector product dataset and convolutional neural nets for estimating the run time and energy consumption of the sparse matrix-vector product. Available at: https:\/\/doi.org\/10.5281\/zenodo.3956057 (accessed July 2020).","DOI":"10.1177\/1094342020953196"},{"key":"bibr3-1094342020953196","unstructured":"Barreda M, Dolz MF, Casta\u00f1o MA (2020b) Convolutional neural nets for estimating the run time and energy consumption of the sparse matrix-vector product-spmv-cnn. Available at: https:\/\/github.com\/hpca-uji\/SpMV-CNN-Model\/releases\/tag\/v1.0.0."},{"key":"bibr4-1094342020953196","doi-asserted-by":"crossref","unstructured":"Barreda M, Dolz MF, Casta\u00f1o MA, et al. (2020c) Performance modeling of the sparse matrix-vector product via convolutional neural networks. The Journal of Supercomputing. Epub ahead of print 04 February 2020. DOI: 10.1007\/s11227-020-03186-1.","DOI":"10.1007\/s11227-020-03186-1"},{"key":"bibr5-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/ICPADS.2016.0120"},{"key":"bibr6-1094342020953196","doi-asserted-by":"publisher","DOI":"10.25080\/Majora-8b375195-004"},{"key":"bibr7-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1145\/1052934.1052938"},{"volume-title":"Pattern Recognition and Machine Learning (Information Science and Statistics)","year":"2006","author":"Bishop CM","key":"bibr8-1094342020953196"},{"key":"bibr9-1094342020953196","first-page":"111","volume-title":"ICML 2010\u2014Proceedings, 27th International Conference on Machine Learning, ICML","author":"Boureau Y","year":"2010"},{"key":"bibr10-1094342020953196","unstructured":"Chollet F. (2015) Keras. Available at: https:\/\/github.com\/fchollet\/keras (accessed July 2020)."},{"key":"bibr11-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1587\/transinf.2017EDP7176"},{"key":"bibr12-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1145\/2049662.2049670"},{"key":"bibr13-1094342020953196","unstructured":"Eijkhout V, Pozo R (1994) Data structures and algorithms for distributed sparse matrix operations. Technical report. University of Tennessee, Knoxville, TN, USA."},{"key":"bibr14-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/ICPP.2017.38"},{"key":"bibr15-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1177\/1094342019842915"},{"key":"bibr16-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2013.43"},{"key":"bibr17-1094342020953196","unstructured":"Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Gordon G, Dunson D, Dudk M (eds) Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, Proceedings of Machine Learning Research, volume 15. Fort Lauderdale, FL, USA: PMLR, pp. 315\u2013323. URL http:\/\/proceedings.mlr.press\/v15\/glorot11a\/glorot11a.pdf (accessed July 2020)."},{"key":"bibr18-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/ScalA.2018.00010"},{"key":"bibr19-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2017.10.013"},{"key":"bibr20-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1515\/9781400830329"},{"key":"bibr21-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"bibr22-1094342020953196","unstructured":"Lee C, Gallagher PW, Tu Z (2015) Generalizing pooling functions in convolutional neural networks: mixed, gated, and tree. CoRR abs\/1509.08985. URL http:\/\/arxiv.org\/abs\/1509.08985 (accessed July 2020)."},{"key":"bibr23-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2014.2308221"},{"key":"bibr24-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-016-1729-4"},{"key":"bibr25-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/ICPPW.2014.30"},{"key":"bibr26-1094342020953196","unstructured":"McCalpin JD (1991\u20132007) Stream: sustainable memory bandwidth in high performance computers. Technical report. Charlottesville, Virginia: University of Virginia. Available at: http:\/\/www.cs.virginia.edu\/stream\/. A continually updated technical report. http:\/\/www.cs.virginia.edu\/stream\/ (accessed July 2020)."},{"key":"bibr27-1094342020953196","doi-asserted-by":"publisher","DOI":"10.29007\/lnnt"},{"key":"bibr28-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS.2014.6844463"},{"key":"bibr29-1094342020953196","unstructured":"Pumperla M (2017) Hyperas: a very simple convenience wrapper around hyperopt for fast prototyping with Keras model. Available at: http:\/\/maxpumperla.com\/hyperas\/ (accessed July 2020)."},{"key":"bibr30-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2014.09.003"},{"key":"bibr31-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2013.73"},{"key":"bibr32-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2012.121"},{"key":"bibr33-1094342020953196","doi-asserted-by":"publisher","DOI":"10.1145\/3200691.3178495"}],"container-title":["The International Journal of High Performance Computing Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/1094342020953196","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/1094342020953196","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/1094342020953196","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T08:22:01Z","timestamp":1740817321000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/1094342020953196"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,8,26]]},"references-count":33,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2021,5]]}},"alternative-id":["10.1177\/1094342020953196"],"URL":"https:\/\/doi.org\/10.1177\/1094342020953196","relation":{},"ISSN":["1094-3420","1741-2846"],"issn-type":[{"type":"print","value":"1094-3420"},{"type":"electronic","value":"1741-2846"}],"subject":[],"published":{"date-parts":[[2020,8,26]]}}}