Write some Software

Problem Description In this assignment, write a Hadoop MapReduce program to output all at once the average, maximum and minimal values of Google cluster trace data version 1 (CPU or Memory usage). The specification of the data format and download link of the data is available from, [login to view URL] The descriptions and analysis of Google trace data can be found in Analysis and Lessons from Publically Available Google Cluster Trace [login to view URL] Google cluster-usage traces: format + schema [login to view URL] A python script is provided to strip 10,000 data records from the original source. You can use the script to generate the sample data sets for your own development and testing. To run this script, python-numpy and python matplotlib module need to be installed. Or you can use the samples provided on the Moodle site. You can choose to code the MapReduce program in Java, C++ or Python. No matter what language you choose to implement, the submission should be packed and executable. ! For Java, all the classes should be packed in a single executable Jar. ! For C++, please provide the make file Pack all your executable in a single archive file. Your program will be tested and run against other slices of the Google trace data.