CSC352 Homework 4 2017

From dftwiki3
Revision as of 12:57, 25 April 2017 by Thiebaut (talk | contribs) (Problem Statement)
Jump to: navigation, search

--D. Thiebaut (talk) 12:23, 25 April 2017 (EDT)


This homework is due the last day of exams 5/12/17, at 4:00 p.m.


Problem Statement


Implement the game of life using Hadoop, on Amazon Web Services.
Please follow these recommendations:

  • Your program should operate on the same data file you used for the MPI version of the Game of Life. The first dish should contain 10 lines of text.
  • Your program should be written in Python (or Java, if you are ambitious).
  • In the header of your program, report on the amount of time taken by the program.
  • Your program needs to compute only 1 generation.
  • You should submit an archive file (zip or tgz extension) on Moodle, and include in it:
  • the source code of your program(s),
  • the original dish as a text file, and
  • the first new generation output by Hadoop as a text file.
  • a task-graph (jpg or pdf) of the execution of the map, reduce and shuffle tasks as a function of time, as illustrated in the third Hadoop lab.


Reference


The program we developed in class for running the Game of Life program from the Linux command line, with python mapper and reducer programs, is available here.