Citation
Adam Cataldo. "Data Flow in the Data Center". Talk or presentation, 7, November, 2013; Presented at the 10th Biennial Ptolemy Miniconference, Berkeley. .

Abstract
In this talk, I will give an overview of Cascading, a data-flow abstraction used to generate MapReduce jobs on Hadoop. I’ll explain how we’re using Cascading to drive analytics at Wealthfront, the world’s largest software-based financial advisor. Then I'll dive into some of the technical problems we had to solve to map Cascading into our workflows. In particular, I'll describe how we move data back and forth between our Cascading cluster and other production systems, how we test our Cascading jobs for correctness, and how we tune for performance to meet real-time constraints.

Electronic downloads

• Cataldo_DataFlowInTheDataCenter_PtolemyMConf2013.pdf · application/pdf · 3018 kbytes

• HTML

  Adam Cataldo. <a
  href="http://chess.eecs.berkeley.edu/pubs/1029.html"><i>Data
  Flow in the Data Center</i></a>, Talk or
  presentation,  7, November, 2013; Presented at the <a
  href="http://ptolemy.org/conferences/13" >10th
  Biennial Ptolemy Miniconference</a>, Berkeley.
  .

• Plain text

  Adam Cataldo. "Data Flow in the Data Center". Talk
  or presentation,  7, November, 2013; Presented at the <a
  href="http://ptolemy.org/conferences/13" >10th
  Biennial Ptolemy Miniconference</a>, Berkeley.
  .

• BibTeX

  @presentation{Cataldo13_DataFlowInDataCenter,
      author = {Adam Cataldo},
      title = {Data Flow in the Data Center},
      day = {7},
      month = {November},
      year = {2013},
      note = {Presented at the <a
                href="http://ptolemy.org/conferences/13" >10th
                Biennial Ptolemy Miniconference</a>, Berkeley.
  },
      abstract = {In this talk, I will give an overview of
                Cascading, a data-flow abstraction used to
                generate MapReduce jobs on Hadoop. I’ll explain
                how we’re using Cascading to drive analytics at
                Wealthfront, the world’s largest software-based
                financial advisor. Then I'll dive into some of the
                technical problems we had to solve to map
                Cascading into our workflows. In particular, I'll
                describe how we move data back and forth between
                our Cascading cluster and other production
                systems, how we test our Cascading jobs for
                correctness, and how we tune for performance to
                meet real-time constraints.},
      URL = {http://chess.eecs.berkeley.edu/pubs/1029.html}
  }
  

Posted by Barb Hoversten on 16 Nov 2013.
Groups: chess
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.