# CS 302 - Data Structures

## Spring 2015

## Programming Assignment 4 : Sorting

## Due on Tuesday, Mar 10 at 9:00 am

Programming Problem 9 on Page 332 (include counting sort and bucket sort)

The object of this programming assignment is twofold. You are to compare the relative performance of different sorting algorithms on the same data set and the same algorithm on different data sets.

Choose a sorting algorithm from each of the O(n^2) (i.e., selection, insertion and buble), O(n logn) (i.e., merge and quick), and O(n) (i.e., counting, bucket and radix) categories.

Randomly generate (1 000, 10 000, 100 000, or 1 000 000) integer values (in the range of 0 to 1000000) to be inserted into the data structures. Note that the list (i.e., the input values) itself should not be sorted and all algorithms should use the same input file when the input size is same.

Bonus: Running 1 000 000 values will be accounted for 2 points bonus.

Also, test the speed of algorithms when the input array is already sorted (for the same input data).

The following output should be provided for an average of 10 sorts of each algorithm and input size:

- the CPU time (note that generation processes should not affect these values)
- the number of comparisons
- the number of swaps

Note: You can report only one run for the ones that take over an hour.

Note: You can only use (i.e., copy) code from the book and slides. You need to develop the rest of the code yourself.

Deliverables: (as source code or a doc or pdf file)

- Source code for your classes (one for each sort algorithm for a total of 3)
- Doxygen reports for code documentation
- A main file that tests the classes with randomly generated data.
- Test output of your codes (timing snapshots)

Teaching Assistant: A. Soran

What to turn in:
A softcopy of your sourcecode and reports should be **uploaded to WebCT**. Do not compress as a zip file or combine as a tar file.