A performance experiment on sequential v.s. MapReduce frameworks using Apache Hadoop. Experiments were run on the Cal Poly on-campus cluster.

Quinn Coleman, Alex Pinto, Logan Anderson [email protected], [email protected], [email protected] Lab 7 - CSC 369 Spring 2020

Problem 2:

• For IowaLiquorSequential, must have Iowa_Liquor_Sales.csv in the same directory

Problem 3:

• Output of Hadoop program "Correlation.java" goes to "./test/CovidCorrelation/"

Data Generation:

• Generate extra files of different size for "Correlation.java" by running "source gen_files.sh". IMPORTANT: In order to make larger files than original, copies of "us-covid-counties.csv" named "ucc-.csv" must be in your cwd. Files are lengthened/shortened by adding/removing lines to the original file.

• Must manually type different input file into "Correlation.java" to use it. "CovidCorrelation.java" has no command line args.