{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T20:51:32Z","timestamp":1765486292265},"reference-count":62,"publisher":"Association for Computing Machinery (ACM)","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2018,12]]},"abstract":"<jats:p>The ability to perform machine learning (ML) tasks in a database management system (DBMS) provides the data analyst with a powerful tool. Unfortunately, integration of ML into a DBMS is challenging for reasons varying from differences in execution model to data layout requirements. In this paper, we assume a column-store main-memory DBMS, optimized for online analytical processing, as our initial system. On this system, we explore the integration of coordinate-descent based methods working natively on columnar format to train generalized linear models. We use a cache-efficient, partitioned stochastic coordinate descent algorithm providing linear throughput scalability with the number of cores while preserving convergence quality, up to 14 cores in our experiments.<\/jats:p>\n          <jats:p>Existing column oriented DBMS rely on compression and even encryption to store data in memory. When those features are considered, the performance of a CPU based solution suffers. Thus, in the paper we also show how to exploit hardware acceleration as part of a hybrid CPU+FPGA system to provide on-the-fly data transformation combined with an FPGA-based coordinate-descent engine. The resulting system is a column-store DBMS with its important features preserved (e.g., data compression) that offers high performance machine learning capabilities.<\/jats:p>","DOI":"10.14778\/3297753.3297756","type":"journal-article","created":{"date-parts":[[2019,2,27]],"date-time":"2019-02-27T14:57:56Z","timestamp":1551279476000},"page":"348-361","source":"Crossref","is-referenced-by-count":29,"title":["ColumnML"],"prefix":"10.14778","volume":"12","author":[{"given":"Kaan","family":"Kara","sequence":"first","affiliation":[{"name":"ETH Zurich, Switzerland"}]},{"given":"Ken","family":"Eguro","sequence":"additional","affiliation":[{"name":"ETH Zurich, Switzerland"}]},{"given":"Ce","family":"Zhang","sequence":"additional","affiliation":[{"name":"Microsoft Research"}]},{"given":"Gustavo","family":"Alonso","sequence":"additional","affiliation":[{"name":"Microsoft Research"}]}],"member":"320","published-online":{"date-parts":[[2018,12]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"https:\/\/github.com\/owenzhang\/Kaggle-AmazonChallenge2013.  https:\/\/github.com\/owenzhang\/Kaggle-AmazonChallenge2013."},{"key":"e_1_2_1_2_1","unstructured":"https:\/\/www.datarobot.com\/blog\/datarobot-the-2014-kdd-cup.  https:\/\/www.datarobot.com\/blog\/datarobot-the-2014-kdd-cup."},{"key":"e_1_2_1_3_1","unstructured":"https:\/\/www.intel.com\/content\/dam\/doc\/white-paper\/advanced-encryption\\-standard-new-instructions-set-paper.pdf.  https:\/\/www.intel.com\/content\/dam\/doc\/white-paper\/advanced-encryption\\-standard-new-instructions-set-paper.pdf."},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1561\/1900000024"},{"key":"e_1_2_1_5_1","volume-title":"CIDR. Citeseer","author":"Arasu A.","year":"2013","unstructured":"A. Arasu , S. Blanas , K. Eguro , R. Kaushik , D. Kossmann , R. Ramamurthy , and R. Venkatesan . Orthogonal Security with Cipherbase . In CIDR. Citeseer , 2013 . A. Arasu, S. Blanas, K. Eguro, R. Kaushik, D. Kossmann, R. Ramamurthy, and R. Venkatesan. Orthogonal Security with Cipherbase. In CIDR. Citeseer, 2013."},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/2723372.2742796"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2915231"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/1989323.1989328"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1137\/16M1080173"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1145\/2463676.2465283"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2018.022071131"},{"key":"e_1_2_1_12_1","first-page":"2674","volume-title":"Advances in neural information processing systems","author":"De Sa C. M.","year":"2015","unstructured":"C. M. De Sa , C. Zhang , K. Olukotun , and C. R\u00e9 . Taming the wild: A unified analysis of hogwild-style algorithms . In Advances in neural information processing systems , pages 2674 -- 2682 , 2015 . C. M. De Sa, C. Zhang, K. Olukotun, and C. R\u00e9. Taming the wild: A unified analysis of hogwild-style algorithms. In Advances in neural information processing systems, pages 2674--2682, 2015."},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/3123939.3123983"},{"issue":"1","key":"e_1_2_1_14_1","first-page":"28","volume":"35","author":"F\u00e4rber F.","year":"2012","unstructured":"F. F\u00e4rber , N. May , W. Lehner , P. Gro\u00dfe , I. M\u00fcller , H. Rauhe , and J. Dees . The SAP HANA Database-An Architecture Overview. IEEE Data Eng. Bull. , 35 ( 1 ): 28 -- 33 , 2012 . F. F\u00e4rber, N. May, W. Lehner, P. Gro\u00dfe, I. M\u00fcller, H. Rauhe, and J. Dees. The SAP HANA Database-An Architecture Overview. IEEE Data Eng. Bull., 35(1):28--33, 2012.","journal-title":"The SAP HANA Database-An Architecture Overview. IEEE Data Eng. Bull."},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2213836.2213874"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2015.46"},{"key":"e_1_2_1_17_1","volume-title":"Why random reshuffling beats stochastic gradient descent. arXiv preprint arXiv:1510.08560","author":"G\u00fcrb\u00fczbalaban M.","year":"2015","unstructured":"M. G\u00fcrb\u00fczbalaban , A. Ozdaglar , and P. Parrilo . Why random reshuffling beats stochastic gradient descent. arXiv preprint arXiv:1510.08560 , 2015 . M. G\u00fcrb\u00fczbalaban, A. Ozdaglar, and P. Parrilo. Why random reshuffling beats stochastic gradient descent. arXiv preprint arXiv:1510.08560, 2015."},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.14778\/2367502.2367510"},{"key":"e_1_2_1_19_1","first-page":"1223","volume-title":"Advances in neural information processing systems","author":"Ho Q.","year":"2013","unstructured":"Q. Ho , J. Cipar , H. Cui , S. Lee , J. K. Kim , P. B. Gibbons , G. A. Gibson , G. Ganger , and E. P. Xing . More effective distributed ml via a stale synchronous parallel parameter server . In Advances in neural information processing systems , pages 1223 -- 1231 , 2013 . Q. Ho, J. Cipar, H. Cui, S. Lee, J. K. Kim, P. B. Gibbons, G. A. Gibson, G. Ganger, and E. P. Xing. More effective distributed ml via a stale synchronous parallel parameter server. In Advances in neural information processing systems, pages 1223--1231, 2013."},{"key":"e_1_2_1_20_1","first-page":"40","article-title":"MonetDB: Two decades of research in column-oriented database architectures","author":"Idreos S.","year":"2012","unstructured":"S. Idreos , F. Groffen , N. Nes , S. Manegold , S. Mullender , and M. Kersten . MonetDB: Two decades of research in column-oriented database architectures . Data Engineering , 40 , 2012 . S. Idreos, F. Groffen, N. Nes, S. Manegold, S. Mullender, and M. Kersten. MonetDB: Two decades of research in column-oriented database architectures. Data Engineering, 40, 2012.","journal-title":"Data Engineering"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.14778\/3137628.3137632"},{"key":"e_1_2_1_22_1","first-page":"3068","volume-title":"Advances in Neural Information Processing Systems","author":"Jaggi M.","year":"2014","unstructured":"M. Jaggi , V. Smith , M. Tak\u00e1c , J. Terhorst , S. Krishnan , T. Hofmann , and M. I. Jordan . Communication-efficient distributed dual coordinate ascent . In Advances in Neural Information Processing Systems , pages 3068 -- 3076 , 2014 . M. Jaggi, V. Smith, M. Tak\u00e1c, J. Terhorst, S. Krishnan, T. Hofmann, and M. I. Jordan. Communication-efficient distributed dual coordinate ascent. In Advances in Neural Information Processing Systems, pages 3068--3076, 2014."},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/2796314.2745867"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2017.39"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2016.7577353"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3035946"},{"key":"e_1_2_1_28_1","volume-title":"Proceedings of the 7th Biennial Conference on Innovative Data Systems Research","author":"Kornacker M.","year":"2015","unstructured":"M. Kornacker : A modern, open-source SQL engine for Hadoop . In Proceedings of the 7th Biennial Conference on Innovative Data Systems Research , 2015 . M. Kornacker et al. Impala: A modern, open-source SQL engine for Hadoop. In Proceedings of the 7th Biennial Conference on Innovative Data Systems Research, 2015."},{"key":"e_1_2_1_29_1","first-page":"2","volume-title":"Cidr","volume":"1","author":"Kraska T.","year":"2013","unstructured":"T. Kraska , A. Talwalkar , J. C. Duchi , R. Griffith , M. J. Franklin , and M. I. Jordan . MLbase: A Distributed Machine-learning System . In Cidr , volume 1 , pages 2 -- 1 , 2013 . T. Kraska, A. Talwalkar, J. C. Duchi, R. Griffith, M. J. Franklin, and M. I. Jordan. MLbase: A Distributed Machine-learning System. In Cidr, volume 1, pages 2--1, 2013."},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3054775"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/2723372.2723713"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.14778\/2824032.2824071"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE.2018.00226"},{"key":"e_1_2_1_34_1","volume-title":"Solid-State Circuits Conference-(ISSCC), 2015","author":"Li P.","year":"2015","unstructured":"P. Li , J. L. Shin , G. Konstadinidis , F. Schumacher , V. Krishnaswamy , H. Cho , S. Dash , R. Masleid , C. Zheng , Y. D. Lin , A 20nm 32-Core 64MB L3 cache SPARC M7 processor. In Solid-State Circuits Conference-(ISSCC), 2015 IEEE International, pages 1--3. IEEE , 2015 . P. Li, J. L. Shin, G. Konstadinidis, F. Schumacher, V. Krishnaswamy, H. Cho, S. Dash, R. Masleid, C. Zheng, Y. D. Lin, et al. 4.2 A 20nm 32-Core 64MB L3 cache SPARC M7 processor. In Solid-State Circuits Conference-(ISSCC), 2015 IEEE International, pages 1--3. IEEE, 2015."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1137\/140961134"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.14778\/3231751.3231770"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.14778\/3236187.3236188"},{"key":"e_1_2_1_38_1","volume-title":"NIPS Workshop on Distributed Machine Learning and Matrix Computations","author":"Noel C.","year":"2014","unstructured":"C. Noel and S. Osindero . Dogwild!-distributed hogwild for cpu & gpu . In NIPS Workshop on Distributed Machine Learning and Matrix Computations , 2014 . C. Noel and S. Osindero. Dogwild!-distributed hogwild for cpu & gpu. In NIPS Workshop on Distributed Machine Learning and Matrix Computations, 2014."},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/ReConFig.2011.4"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.23919\/FPL.2017.8056784"},{"key":"e_1_2_1_41_1","first-page":"1","volume-title":"Hot Chips 26 Symposium (HCS), 2014","author":"Putnam A.","year":"2014","unstructured":"A. Putnam . Large-scale reconfigurable computing in a Microsoft datacenter . In Hot Chips 26 Symposium (HCS), 2014 IEEE, pages 1 -- 38 . IEEE, 2014 . A. Putnam. Large-scale reconfigurable computing in a Microsoft datacenter. In Hot Chips 26 Symposium (HCS), 2014 IEEE, pages 1--38. IEEE, 2014."},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/2678373.2665678"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2018.00016"},{"key":"e_1_2_1_44_1","first-page":"11","volume-title":"Conference on Learning Theory","author":"Recht B.","year":"2012","unstructured":"B. Recht and C. R\u00e9 . Toward a noncommutative arithmetic-geometric mean inequality: conjectures, case-studies, and consequences . In Conference on Learning Theory , pages 11 -- 11 , 2012 . B. Recht and C. R\u00e9. Toward a noncommutative arithmetic-geometric mean inequality: conjectures, case-studies, and consequences. In Conference on Learning Theory, pages 11--1, 2012."},{"key":"e_1_2_1_45_1","first-page":"693","volume-title":"Advances in neural information processing systems","author":"Recht B.","year":"2011","unstructured":"B. Recht , C. Re , S. Wright , and F. Niu . Hogwild: A lock-free approach to parallelizing stochastic gradient descent . In Advances in neural information processing systems , pages 693 -- 701 , 2011 . B. Recht, C. Re, S. Wright, and F. Niu. Hogwild: A lock-free approach to parallelizing stochastic gradient descent. In Advances in neural information processing systems, pages 693--701, 2011."},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-015-0901-6"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/CCECE.2007.315"},{"key":"e_1_2_1_48_1","volume-title":"Stochastic methods for l1-regularized loss minimization. Journal of Machine Learning Research, 12(Jun):1865--1892","author":"Shalev-Shwartz S.","year":"2011","unstructured":"S. Shalev-Shwartz and A. Tewari . Stochastic methods for l1-regularized loss minimization. Journal of Machine Learning Research, 12(Jun):1865--1892 , 2011 . S. Shalev-Shwartz and A. Tewari. Stochastic methods for l1-regularized loss minimization. Journal of Machine Learning Research, 12(Jun):1865--1892, 2011."},{"key":"e_1_2_1_49_1","first-page":"46","volume-title":"Advances in Neural Information Processing Systems","author":"Shamir O.","year":"2016","unstructured":"O. Shamir . Without-replacement sampling for stochastic gradient methods . In Advances in Neural Information Processing Systems , pages 46 -- 54 , 2016 . O. Shamir. Without-replacement sampling for stochastic gradient methods. In Advances in Neural Information Processing Systems, pages 46--54, 2016."},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3035954"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3058746"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3064043"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/2370816.2370874"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1007\/0-387-25465-X_63"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.2517-6161.1996.tb02080.x"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021744"},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.14778\/2732967.2732976"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.14778\/2732977.2733001"},{"key":"e_1_2_1_60_1","first-page":"1719","volume-title":"Advances in Neural Information Processing Systems","author":"Zhang Z.","year":"2017","unstructured":"Z. Zhang and M. Brand . Convergent block coordinate descent for training tikhonov regularized deep neural networks . In Advances in Neural Information Processing Systems , pages 1719 -- 1728 , 2017 . Z. Zhang and M. Brand. Convergent block coordinate descent for training tikhonov regularized deep neural networks. In Advances in Neural Information Processing Systems, pages 1719--1728, 2017."},{"key":"e_1_2_1_61_1","first-page":"3329","volume-title":"Advances in neural information processing systems","author":"Zhao T.","year":"2014","unstructured":"T. Zhao , M. Yu , Y. Wang , R. Arora , and H. Liu . Accelerated mini-batch randomized block coordinate descent method . In Advances in neural information processing systems , pages 3329 -- 3337 , 2014 . T. Zhao, M. Yu, Y. Wang, R. Arora, and H. Liu. Accelerated mini-batch randomized block coordinate descent method. In Advances in neural information processing systems, pages 3329--3337, 2014."},{"key":"e_1_2_1_62_1","volume-title":"On the convergence properties of a K-step averaging stochastic gradient descent algorithm for nonconvex optimization. arXiv preprint arXiv:1708.01012","author":"Zhou F.","year":"2017","unstructured":"F. Zhou and G. Cong . On the convergence properties of a K-step averaging stochastic gradient descent algorithm for nonconvex optimization. arXiv preprint arXiv:1708.01012 , 2017 . F. Zhou and G. Cong. On the convergence properties of a K-step averaging stochastic gradient descent algorithm for nonconvex optimization. arXiv preprint arXiv:1708.01012, 2017."}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3297753.3297756","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T09:35:08Z","timestamp":1672220108000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3297753.3297756"}},"subtitle":["column-store machine learning with on-the-fly data transformation"],"short-title":[],"issued":{"date-parts":[[2018,12]]},"references-count":62,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2018,12]]}},"alternative-id":["10.14778\/3297753.3297756"],"URL":"https:\/\/doi.org\/10.14778\/3297753.3297756","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2018,12]]}}}