@inproceedings{DhillonModha2000, author = {Dhillon, Inderjit S. and Modha, Dharmendra S.}, title = {A Data-Clustering Algorithm on Distributed Memory Multiprocessors}, booktitle = {Revised Papers from Large-Scale Parallel Data Mining, Workshop on Large-Scale Parallel KDD Systems, SIGKDD}, year = {2000}, isbn = {3-540-67194-3}, pages = {245--260}, publisher = {Springer-Verlag}, address = {London, UK}, } @InProceedings{AzureKMeans, author = {Durut, Matthieu and Rossi, Fabrice}, title = {K-means on Azure}, booktitle = {LCCC: NIPS 2010 Workshop on Learning on Cores, Clusters and Clouds}, year = 2010, address = {Whistler, Canada}, month = {December}} @InProceedings{DurutRossiEsann2011, author = {Durut, Matthieu and Rossi, Fabrice}, title = {Communication Challenges in Cloud K-means}, booktitle = {Proceedings of XIXth European Symposium on Artificial Neural Networks (ESANN 2011)}, address = {Bruges, Belgique}, pages = {387--392}, year = 2011, month = {April}} @inproceedings{Graphlab2010, title = {GraphLab: A New Parallel Framework for Machine Learning}, author = {Yucheng Low and Joseph Gonzalez and Aapo Kyrola and Danny Bickson and Carlos Guestrin and Joseph M. Hellerstein}, booktitle = {Conference on Uncertainty in Artificial Intelligence (UAI)}, month = {July}, year = {2010}, address = {Catalina Island, California}, wwwfilebase = {uai2010-low-gonzalez-kyrola-bickson-guestrin-hellerstein}, wwwtopic = {Parallel Learning}, } @inproceedings{MapReduce2004, author = {Dean, Jeffrey and Ghemawat, Sanjay}, title = {MapReduce: simplified data processing on large clusters}, booktitle = {OSDI'04: Proceedings of the 6th conference on Symposium on Opearting Systems Design \& Implementation}, year = {2004}, pages = {10--10}, location = {San Francisco, CA}, publisher = {USENIX Association}, address = {Berkeley, CA, USA}, file = {hpc/google/mapreduce-osdi04.pdf}, } @Article{KMeansJoshi, title = {Parallel K-means Algorithm on Distributed Memory Multiprocessors}, author = {Manasi N.Joshi}, journal = {?}, year = {2003}, month = {spring}, note* = {un papier sur un K-Means distribué} } @Article{StragglerIssuesInMapReduce, title = {The Curse of Zipf and Limits to Parallelization: A Look at the Stragglers Problem in MapReduce}, author = {Jimmy Lin}, note* = {Discussion on the straggler issue, good description of why is there an issue, etc... Very naive solution but the interest is explaining rather than proposing a solution} } @ARTICLE{TSI1, author = { Tsitsiklis, J. and Bertsekas, D. and Athans, M.}, journal = {Automatic Control, IEEE Transactions on}, number = {9}, pages = {803-812}, title = {Distributed asynchronous deterministic and stochastic gradient optimization algorithms}, volume = {31}, year = {1986}, issn = {0018-9286}, month = Sep, abstract = {We present a model for asynchronous distributed computation and then proceed to analyze the convergence of natural asynchronous distributed versions of a large class of deterministic and stochastic gradient-like algorithms. We show that such algorithms retain the desirable convergence properties of their centralized counterparts, provided that the time between consecutive interprocessor communications and the communication delays are not too large.} Annote = {Fundamental paper that leads me for building asynchronous models. This paper have been reedited in the Tsitsiklis MIT book. This article proves the converbence of an asynchronous iteration of the form $$w_i(t+1)= w_i(t) + \varepislon_t\left(\frac{\partial C}{\partial w}(w^i(t) + \ksi_t) \right)$$ where $\ksi_t$ is a white noise. There is also something else the model have been somewhat complicated by introducing a competition between the processors (different processors can update the same composent)} } @incollection{LAN1, author = {Zinkevich, Martin and Smola, Alex and Langford, John}, booktitle = {Advances in Neural Information Processing Systems 22}, citeulike-article-id = {6962859}, editor = {Bengio, Y. and Schuurmans, D. and Lafferty, J. and Williams, C. K. I. and Culotta, A.}, keywords = {learning, online}, pages = {2331--2339}, posted-at = {2010-04-06 15:38:04}, priority = {2}, title = {Slow Learners are Fast}, year = {2009} } @ARTICLE{LOU1, AUTHOR = {Gilles Louppe, Pierre Geurts}, TITLE = {A zealous parallel gradient descent algorithm}, JOURNAL = {NIPS Workshop}, YEAR = {2010}, }