Project 3: Hash Tables!

Due: Monday 10/26/2009

For this assignment you will implement a superTable data structure to store a collection of students with integer id numbers and double gpa scores. The superTable must implement the following methods:

class SuperTable

{

public:

//add new student with given id and gpa to data structure

void insertStudent( unsigned int id, double gpa );

//return the gpa of the student with given id.

//Return -1 if student is not in data structure

double getGPA( unsigned int id );

//change the gpa of the student with the given id

//to newGPA.

void updateGPA( unsigned int id, double newGPA );

};

Further, your superTable must consist of 2 sub-data structures. The first is a hash table of students that resolves collisions via chaining. The second is a Bloom filter. The Bloom filter should be used to speed up the getGPA method for the cases in which the requested student is not in the hashtable, and the insertStudent method in the cases in which the student is already in the table.

You must design your data structure so that it can quickly deal with all queries even if there are 1,000,000 students in the data structure. You should experiment with different size inputs. How big should your hash table and Bloom filter be to deal with different numbers of students n? How many hash functions k should you use for your Bloom filter? Include a 1-page write up analyzing the speed of different Bloom filter and Hash table sizes, along with the number of hash functions used.