Accelerating Generalized Linear Models with MLWeaving: A One-Size-Fits-All System for Any-precision Learning

  title={Accelerating Generalized Linear Models with MLWeaving: A One-Size-Fits-All System for Any-precision Learning},
  author={Z. Wang and Kaan Kara and H. Zhang and G. Alonso and Onur Mutlu and Ce Zhang},
  journal={Proc. VLDB Endow.},
  • Z. Wang, Kaan Kara, +3 authors Ce Zhang
  • Published 2019
  • Computer Science, Mathematics
  • Proc. VLDB Endow.
  • Learning from the data stored in a database is an important function increasingly available in relational engines. Methods using lower precision input data are of special interest given their overall higher efficiency but, in databases, these methods have a hidden cost: the quantization of the real value into a smaller number is an expensive step. To address the issue, in this paper we present MLWeaving, a data structure and hardware acceleration technique intended to speed up learning of… CONTINUE READING
    12 Citations
    doppioDB 2.0: Hardware Techniques for Improved Integration of Machine Learning into Databases
    • 4
    • PDF
    BiS-KM: Enabling Any-Precision K-Means on FPGAs
    • 3
    doppioDB 1.0: Machine Learning inside a Relational Engine
    • 8
    • PDF
    Lowering the Latency of Data Processing Pipelines Through FPGA based Hardware Acceleration
    • 14
    • PDF
    Distributed Inference over Decision Tree Ensembles on Clusters of FPGAs
    • 2
    High Bandwidth Memory on FPGAs: A Data Analytics Perspective
    • 1
    • PDF
    Shuhai: Benchmarking High Bandwidth Memory On FPGAS
    • 7
    • PDF
    Benchmarking High Bandwidth Memory on FPGAs
    • 1
    • PDF
    PipeArch: Generic and Context-Switch Capable Data Processing on FPGAs


    TABLA: A unified template-based framework for accelerating statistical machine learning
    • 107
    • PDF
    ColumnML: Column-Store Machine Learning with On-The-Fly Data Transformation
    • 17
    • PDF
    In-RDBMS Hardware Acceleration of Advanced Analytics
    • 23
    • PDF
    Stripes: Bit-serial deep neural network computing
    • 214
    • Highly Influential
    • PDF
    FPGA-Accelerated Dense Linear Machine Learning: A Precision-Convergence Trade-Off
    • 46
    • PDF
    doppioDB: A hardware accelerated database
    • 12
    DaDianNao: A Machine-Learning Supercomputer
    • Yunji Chen, T. Luo, +8 authors O. Temam
    • Computer Science
    • 2014 47th Annual IEEE/ACM International Symposium on Microarchitecture
    • 2014
    • 884
    • PDF
    Centaur: A Framework for Hybrid CPU-FPGA Databases
    • 46
    • PDF
    Compressed Linear Algebra for Large-Scale Machine Learning
    • 29
    • PDF
    Minerva: Enabling Low-Power, Highly-Accurate Deep Neural Network Accelerators
    • 196
    • PDF