Petuum: A New Platform for Distributed Machine Learning on Big Data

@article{Xing2015PetuumAN,
  title={Petuum: A New Platform for Distributed Machine Learning on Big Data},
  author={E. Xing and Q. Ho and Wei Dai and J. Kim and Jinliang Wei and S. Lee and X. Zheng and Pengtao Xie and Abhimanu Kumar and Y. Yu},
  journal={IEEE Transactions on Big Data},
  year={2015},
  volume={1},
  pages={49-67}
}
  • E. Xing, Q. Ho, +7 authors Y. Yu
  • Published 2015
  • Computer Science
  • IEEE Transactions on Big Data
  • What is a systematic way to efficiently apply a wide spectrum of advanced ML programs to industrial scale problems, using Big Models (up to 100 s of billions of parameters) on Big Data (up to terabytes or petabytes)? Modern parallelization strategies employ fine-grained operations and scheduling beyond the classic bulk-synchronous processing paradigm popularized by MapReduce, or even specialized graph-based execution that relies on graph representations of ML programs. The variety of approaches… CONTINUE READING
    205 Citations
    A Collective Communication Layer for the Software Stack of Big Data Analytics
    • Bingjing Zhang
    • Computer Science
    • 2016 IEEE International Conference on Cloud Engineering Workshop (IC2EW)
    • 2016
    • 1
    • PDF
    Strategies and Principles of Distributed Machine Learning on Big Data
    • 67
    • PDF
    Collaborative Filtering as a Case-Study for Model Parallelism on Bulk Synchronous Systems
    • 2
    • Highly Influenced
    • PDF
    Towards Production-Grade, Platform-Independent Distributed ML
    • 4
    • PDF
    Litz: Elastic Framework for High-Performance Distributed Machine Learning
    • 29
    • PDF
    SparCML: high-performance sparse communication for machine learning
    • 45
    • PDF
    Towards MapReduce based Bayesian deep learning network for monitoring big data applications
    • 4
    MapReduce Based Classification for Fault Detection in Big Data Applications
    • M. Shafiq, M. Fekri, R. Ibrahim
    • Computer Science
    • 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)
    • 2017

    References

    SHOWING 1-7 OF 7 REFERENCES
    Distributed GraphLab : A Framework for Machine Learning and Data Mining in the Cloud
    • 1,249
    • Highly Influential
    • PDF
    Spark: Cluster Computing with Working Sets
    • 4,369
    • Highly Influential
    • PDF
    Scaling Distributed Machine Learning with the Parameter Server
    • 1,048
    • Highly Influential
    • PDF
    Hadoop: The Definitive Guide
    • 3,881
    • Highly Influential
    • PDF
    Feature Clustering for Accelerating Parallel Coordinate Descent
    • 71
    • Highly Influential
    • PDF
    ImageNet: A large-scale hierarchical image database
    • 6,950
    • Highly Influential
    • PDF
    Parallel Coordinate Descent for L1-Regularized Loss Minimization
    • 296
    • Highly Influential
    • PDF