Building a Distributed, GPU-based Machine Learning Library

James Jia, Pradeep Kalipatnapu and Yiheng Yang

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2016-112
May 17, 2016

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-112.pdf

BIDData is a machine learning library under development at Berkeley, whose use of the GPU gives it a single core performance that outpaces its peers. My capstone team and I have integrated BIDData with Apache Spark to create a distributed, GPU-based machine learning library for future big data applications.


BibTeX citation:

@techreport{Jia:EECS-2016-112,
    Author = {Jia, James and Kalipatnapu, Pradeep and Yang, Yiheng},
    Editor = {Canny, John F.},
    Title = {Building a Distributed, GPU-based Machine Learning Library},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2016},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-112.html},
    Number = {UCB/EECS-2016-112},
    Abstract = {BIDData is a machine learning library under development at Berkeley, whose use of the GPU gives it a single core performance that outpaces its peers. My capstone team and I have integrated BIDData with Apache Spark to create a distributed, GPU-based machine learning library for future big data applications.}
}

EndNote citation:

%0 Report
%A Jia, James
%A Kalipatnapu, Pradeep
%A Yang, Yiheng
%E Canny, John F.
%T Building a Distributed, GPU-based Machine Learning Library
%I EECS Department, University of California, Berkeley
%D 2016
%8 May 17
%@ UCB/EECS-2016-112
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-112.html
%F Jia:EECS-2016-112