Division-remainder hashing, Data Structure & Algorithms

According to this, key value is divided by any fitting number, generally a prime number, and the division of remainder is utilized as the address for the record.

The choice of suitable divisor may not be so simple. If it is known that the file is to have n records, then we has to, assuming that only one record can be stored at a given address, have divisor n.

Also we might have a very large key space as compared to the address space. Key space refers to every possible key value. The address space possibly may not match in fact for the key values in the file. So, calculated address may not be unique. It is called Collision, that means

R(K1) = R(K2), where K1 = K2.

Two unequal keys have been calculated to have the simialr address. The keys are called as synonyms.

There are several approaches to handle the problem of collisions. One of these is to hash to buckets. A bucket is a space that can suggest multiple records.

Posted Date: 4/11/2013 6:23:25 AM | Location : United States







Related Discussions:- Division-remainder hashing, Assignment Help, Ask Question on Division-remainder hashing, Get Answer, Expert's Help, Division-remainder hashing Discussions

Write discussion on Division-remainder hashing
Your posts are moderated
Related Questions

1)    The set of the algorithms whose order is O (1) would run in the identical time.  True/False 2)    Determine the complexity of the following program into big O notation:

Q.1 Compare two functions n 2 and 2 n for various values of n. Determine when second becomes larger than first. Q.2 Why do we use asymptotic notation in the study of algorit

The below figure illustrates the BOM (Bill of Materials) for product A. The MPS (Material requirements Planning) start row in the master production schedule for product A calls for

In what ways we can get the context sensitive F1 help on a field?' Data element documentation. Data element additional text in screen painter. Using the process on help r

Q. Implement a stack making use of the linked list. Show the PUSH and POP operations both. A n s . Stack implemantation using linked list # include # include

Program: Program segment for insertion of an element into the queue add(int value) { struct queue *new; new = (struct queue*)malloc(sizeof(queue)); new->value = val

Searching is the procedure of looking for something: Finding one piece of data that has been stored inside a whole group of data. It is frequently the most time-consuming part of m

Taking a suitable example explains how a general tree can be shown as a Binary Tree. Conversion of general trees to binary trees: A general tree can be changed into an equiv

Multilist Representation of graph