Cloud Computing for Data Analysis:

Spark Apriori Algorithm Execution Steps using DSBA-HADOOP Cluster:

1. Download the .zip file from http://webpages.uncc.edu/aatzache/ITCS6190/Exercises/07_Exercise_SparkAssociationRules.zip

2. Extract the .zip file

3. This .zip file contains the source code(SparkAssociationRules.zip), the executable 'SparkAssociationRules.jar' file and data files inside the 'Data' folder

4. Copy the 'SparkAssociationRules.jar' file to the cluster using WinSCP or CyberDuck or SSH

5. Similarly, copy the Data CarData.txt to the cluster using WinSCP or CyberDuck or SSH

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

7. Copy the data file 'CarData.txt' into HDFS using the following command:
	hadoop fs -put /users/UNCC_USERNAME/CarData.txt /user/UNCC_USERNAME/

8. To execute the program, use the following command:
	spark2-submit --class org.Apriori.Driver --master yarn --deploy-mode client /users/UNCC_USERNAME/SparkAssociationRules.jar /user/UNCC_USERNAME/CarData.txt MINIMUM_SUPPORT_VALUE MINIMUM_CONFIDENCE_VALUE /user/UNCC_USERNAME/Apriori_Output
***NOTE: 
	Give MINIMUM_SUPPORT_VALUE as an integer value
	Give MINIMUM_CONFIDENCE_VALUE as an integer or a float value

9. Get the output using the following command:
	IF THE OUTPUT IS IN SINGLE FILE, USE:
	hadoop fs -get /user/UNCC_USERNAME/Apriori_Output/part-00000 /users/UNCC_USERNAME/Apriori_Output.txt

	IF THE OUTPUT IS IN MULTIPLE FILES, USE:
	hadoop fs -getmerge /user/UNCC_USERNAME/Apriori_Output/ /users/UNCC_USERNAME/Apriori_Output.txt
	
10. Execute the program for both CarData.txt and SampleData.txt
	
11. Upload Apriori_Output.txt along with all commands, results and errors if any from the terminal windows in Canvas