{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,17]],"date-time":"2026-03-17T18:29:06Z","timestamp":1773772146490,"version":"3.50.1"},"reference-count":10,"publisher":"Association for Computing Machinery (ACM)","issue":"1-2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2010,9]]},"abstract":"<jats:p>This paper describes a high performance sampling architecture for inference of latent topic models on a cluster of workstations. Our system is faster than previous work by over an order of magnitude and it is capable of dealing with hundreds of millions of documents and thousands of topics.<\/jats:p>\n          <jats:p>The algorithm relies on a novel communication structure, namely the use of a distributed (key, value) storage for synchronizing the sampler state between computers. Our architecture entirely obviates the need for separate computation and synchronization phases. Instead, disk, CPU, and network are used simultaneously to achieve high performance. We show that this architecture is entirely general and that it can be extended easily to more sophisticated latent variable models such as n-grams and hierarchies.<\/jats:p>","DOI":"10.14778\/1920841.1920931","type":"journal-article","created":{"date-parts":[[2014,6,24]],"date-time":"2014-06-24T12:17:57Z","timestamp":1403612277000},"page":"703-710","source":"Crossref","is-referenced-by-count":232,"title":["An architecture for parallel topic models"],"prefix":"10.14778","volume":"3","author":[{"given":"Alexander","family":"Smola","sequence":"first","affiliation":[{"name":"Yahoo! Research, Santa Clara, CA, and Australian National University, Canberra"}]},{"given":"Shravan","family":"Narayanamurthy","sequence":"additional","affiliation":[{"name":"Yahoo! Labs, Bangalore, India"}]}],"member":"320","published-online":{"date-parts":[[2010,9]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1109\/18.825794"},{"key":"e_1_2_1_2_1","first-page":"81","volume-title":"NIPS","author":"Asuncion A.","year":"2008","unstructured":"A. Asuncion , P. Smyth , and M. Welling . Asynchronous distributed learning of topic models . In NIPS , pages 81 -- 88 . MIT Press , 2008 . A. Asuncion, P. Smyth, and M. Welling. Asynchronous distributed learning of topic models. In NIPS, pages 81--88. MIT Press, 2008."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.5555\/944919.944937"},{"key":"e_1_2_1_4_1","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511804441","volume-title":"Convex Optimization","author":"Boyd S.","year":"2004","unstructured":"S. Boyd and L. Vandenberghe . Convex Optimization . Cambridge University Press , UK , 2004 . S. Boyd and L. Vandenberghe. Convex Optimization. Cambridge University Press, UK, 2004."},{"key":"e_1_2_1_5_1","volume-title":"AISTATS","author":"Gonzalez J.","year":"2009","unstructured":"J. Gonzalez , Y. Low , and C. Guestrin . Residual splash for optimally parallelizing belief propagation . In AISTATS , Clearwater Beach, FL , 2009 . J. Gonzalez, Y. Low, and C. Guestrin. Residual splash for optimally parallelizing belief propagation. In AISTATS, Clearwater Beach, FL, 2009."},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0307752101"},{"key":"e_1_2_1_7_1","volume-title":"NIPS","author":"Newman D.","year":"2009","unstructured":"D. Newman , A. Asuncion , P. Smyth , and M. Welling . Distributed algorithms for topic models , NIPS 2009 . D. Newman, A. Asuncion, P. Smyth, and M. Welling. Distributed algorithms for topic models, NIPS 2009."},{"key":"e_1_2_1_8_1","first-page":"1973","volume-title":"NIPS","author":"Wallach H.","year":"2009","unstructured":"H. Wallach , D. Mimno , and A. McCallum . Rethinking LDA: Why priors matter . NIPS , p. 1973 -- 1981 . 2009 . H. Wallach, D. Mimno, and A. McCallum. Rethinking LDA: Why priors matter. NIPS, p. 1973--1981. 2009."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-02158-9_26"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1145\/1557019.1557121"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/1920841.1920931","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T11:35:57Z","timestamp":1672227357000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/1920841.1920931"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2010,9]]},"references-count":10,"journal-issue":{"issue":"1-2","published-print":{"date-parts":[[2010,9]]}},"alternative-id":["10.14778\/1920841.1920931"],"URL":"https:\/\/doi.org\/10.14778\/1920841.1920931","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2010,9]]}}}