Cloud Computing for Data Analysis

Group Activity 03 - Part 2 in DSBA-Hadoop cluster

1. Download package from https://webpages.uncc.edu/aatzache/ITCS6190/Exercises/GroupActivity03_Part2_FPGrowth.zip

2. This .zip file contains source code (.zip file), Data (in the Data folder) and an executable .jar file

3. Extract GroupActivity03_FPGrowth.zip file and copy .jar file and data files to the DSBA-cluster using WinSCP or CyberDuck or SSH

4. Login to the DSBA-cluster using Putty (Windows) or Terminal (Linux/MAC)

5. Copy the data files to HDFS using following commands:
	hadoop fs -put FILEPATH/FILENAME /user/UNCC_USERNAME/
	
6. To execute the program, use following command:
	spark2-submit --class org.FP.Driver --master yarn --deploy-mode client JAR_FILEPATH /user/UNCC_USERNAME/FILENAME MINIMUM_SUPPORT MINIMUM_CONFIDENCE OUTPUT_PATH
	
	***NOTE:
		Give MINIMUM_SUPPORT MINIMUM_CONFIDENCE values as DOUBLE values (eg) 0.4 or 0.75
		
7. Get output uisng following commands:
	hadoop fs -get OUTPUT_PATH/part-00000 /users/UNCC_USERNAME/FP_growth_Output.txt
	
8. Execute the program for both data (SampleData.txt and ExerciseData.txt attached in the .zip file)

9. Upload all output files, all commands, results and errors, if any, from the terminal window to Canvas