Description

CSCI312 Big Data Management
Assignment 1

Scope
The objectives of Assignment 1 include implementation of HDFS applications, implementation of simple MapReduce applications, and describing an implementation of complex MapReduce application.

This assignment is worth 10% of the total evaluation in the subject.

The assignment consists of 4 tasks and specification of each task starts from a new page.

Only electronic submission through Moodle at:
https://moodle.uowplatform.edu.au/login/index.php
will be accepted. A submission procedure is explained at the end of Assignment 1 specification.

Only one submission of Assignment 1 is allowed and only one submission per student is accepted.

A submission marked by Moodle as “late” is always treated as a late submission no matter how many seconds it is late.

A submission that contains an incorrect file attached is treated as a correct submission with all consequences coming from the evaluation of the file attached.

All files left on Moodle in a state “Draft(not submitted)” will not be evaluated.

A submission of compressed files (zipped, gzipped, rared, tared, 7-zipped, lhzed, … etc) is not allowed. The compressed files will not be evaluated.

Task 1 (2 marks)
Implementation of HDFS application

This task is based on the source code included in a presentation 04 HDFS Interfaces.

Implement in Java HDFS application, that can be used to move a file from one location in HDFS into another location in HDSF.

The application must have the following two parameters.
(1) A path to and a name of file to be moved from.
(2) A path to and a new name of file to be moved to.

Perform the following steps.

(1) Implement the application and save its source code in .java file. A name of file is up to you.

(2) Compile the Java source code and create a jar file.

(3) Upload to HDFS a small text file for the purpose of future testing. A name and location of the file in HDFS is up to you.

(4) Use Hadoop to process your application that moves a file on HDFS from one location to the other.

(5) Use Hadoop to provide an evidence that the file earlier uploaded to HDSF has been successfully moved.

Deliverables
A file solution1.java with a source code of the application that moves a file in HDFS. A file solution1.pdf that contains the contents of Terminal window with a report from compilation, creation of jar file, uploading to HDFS two small files for testing, processing of the application, and an evidence that two files uploaded into HDFS has been successful merges in one file in HDFS.

Task 2 (3 marks)
Implementation of MapReduce application

Assume, that The Bureau of Meteorology records total yearly rainfall in a number of cities located in different states. The measurements are recorded in a text file, that contains data from a period of the last year.

For example, a sample file with the recorded amounts of rainfall could be the following. The first column contains a name of a state, the second column contains a name of a city in a state and the third column contains total rainfall depth per year measured in mm.

Queensland Gold Coast 25
Victoria Melbourne 125
Victoria Geelong 90
Victoria Wodonga 10
NSW Lismore 900
Queensland Brisbane 50
South Australia Adelaide 300
Western Australia Perth 200
Western Australia Albany 200
Western Australia Broome 10

Your task is to implement a MapReduce application, that finds the total rainfall in each state, the largest rainfall in one location in each state and the smallest rainfall in one location in each state.

For example, your application should generate the following outputs when processing data listed above.

Queensland 75 50 25
Victoria 225 125 10
NSW 900 900 900
South Australia 300 300 300
Western Australia 410 200 10

An input file with the speed measurements must include 10 lines listed above and it must contain at least 10 other measurements. All additional measurements are up to you.

Save your solution in a file solution2.java.

When ready, compile, create jar file, and process your application. Display the results created by the application. Next, list your input file with the speed measurements. When finished, Copy and Paste the messages from a Terminal screen into a file solution2.pdf.

Deliverables
A file solution2.java with a source code of the application that implement the functionality of SELECT statement given above. A file solution2.pdf with a report from compilation, creating jar file, processing, displaying the results of processing solution2.java, and listing of your input file with the rainfall measurements.

Task 3 (2 marks)
Implementation of MapReduce application

Consider a classical MapReduce application that counts the total number of occurrences of words in a given text. For example, look at WordCount application available in a file WordCount.java in Laboratory 2.

Assume the following classification of words depending on the length of each word.

very short: 1 <= length <= 3 short: 4 <= length <= 5 medium: 6 <= length <= 8 long: 9 <= length <= 12 X long: 13 <= length <= 15
XX long: 16 <= length

Extend Java code of the application such that it counts in a given text the total number of words in each category. For example, distribution of words in a text that consists of 90 words could be the following.

X short: 10 words short: 15 words medium: 35 words long: 20 words X long: 10 words
XX long: 0 words

Save your solution in a file solution3.java.

When ready, compile, create jar file, and process your application. To test your application, you can use a file sales.txt included in a folder with a specification of Exercise 2. Display the results created by the application. When finished, Copy and Paste the messages from a Terminal screen into a file solution3.pdf.

Deliverables
A file solution3.java with a source code of the application that implement the functionality of SELECT statement given above. A file solution3.pdf with a report from compilation, creating jar file, processing, and displaying the results of processing solution3.java.

Task 4 (3 marks)
Describing MapReduce implementation

Assume, that a text file crime-stories.txt contain the texts of large number of crime stories. Assume, that the file is formatted such that one statement is located in one line of the text file.

Assume, that a text file patterns.txt contains the text patterns, for example regular expressions. Assume, that the file is formatted, such that one pattern is located in one line of the text file.

To simplify the problem, assume that all text patterns in a file patterns.txt are different.

Finally, assume that a function match(text-line, text-pattern) returns true when text-line matches a pattern text-pattern. Otherwise, the function returns false.

Your task is to explain how to implement a MapReduce application, that for each text pattern in a file patterns.txt finds the total number of statements in a file crime-stories.txt that match the pattern.

You must specify the parameters (if any) of your application and the key-value data in the input and output of the Map and Reduce stages
.
There is no need to write Java code, however, if you like it then it is all right to do so. The precise explanations in plain English or in a pseudocode will do. Please note, that if you decide to use pseudocode then your explanations must precisely explain what happens at each stage of Map-Reduce application.

Save you explanations in a file solution4.pdf. This task does not require you to write any code in Java. However, the comprehensive explanations related to all stages of data processing are expected. You are allowed to support your explanations with the fragments of pseudocode. Try to be as specific as it is possible.

Deliverables
A file solution4.pdf with the comprehensive explanations how would you implement in Java a MapReduce application that for each text pattern in a file patterns.txt finds the total number of statements in a file crimestories.txt that match the pattern.

Submission of Assignment 1

Note, that you have only one submission. So, make it absolutely sure that you submit the correct files with the correct contents. No other submission is possible !

Submit the files solution1.java, solution1.pdf, solution2.java, solution2.pdf, solution3.java, solution3.pdf, and solution4.pdf through Moodle in the following way:
(1) Access Moodle at http://moodle.uowplatform.edu.au/
(2) To login use a Login link located in the right upper corner the Web page or in the middle of the bottom of the Web page
(3) When logged select a site ISIT312 (SP222) Big Data Management
(4) Scroll down to a section ASSESSMENT ITEMS (ASSIGNMENTS)
(5) Click at In this place you can submit the outcomes of your work on the tasks included in Assignment 1 link.
(6) Click at a button Add Submission
(7) Move a file solution1.java into an area You can drag and drop files here to add them. You can also use a link Add…
(8) Repeat step (7) for the remaining files solution1.pdf, solution2.java, solution2.pdf, solution3.java, solution3.pdf, and solution4.pdf
(9) Click at a button Save changes
(10) Click at a button Submit assignment
(11) Click at the checkbox with a text attached: By checking this box, I confirm that this submission is my own work, … in order to confirm authorship of your submission.
(12) Click at a button Continue

End of specification