Sorting using interconnection networks, Computer Networking

Assignment Help:

The combinational circuits use the comparators for comparing the storing and number them on the basis of minimum and maximum functions. Likewise, in the interconnection networks the two processors perform the computation of minimum and maximum functions in the following way:

Let us consider there are two processors pj and pi. Each of these processors has been given as input a part of the sequence, say ej and ei. Now, the processor pi sends the element ei to pj and processor pj sends ej to pi. After that, processor pi determine the minimum of ei and ej i.e., min (ei,ej) and processor pj determine the maximum of ei and ej, i.e. max (ei,ej). The above method is called as compare-exchange and it has been depicted in the Figure .

581_Illustration of Exchange-cum-Comparison in interconnection networks.png

                                                          Illustration of Exchange-cum-Comparison in interconnection networks

 The sorting problem chosen is bubble sort and the interconnection network can be depicted as n processors interconnected with each other in the type of a linear array as given in Figure. The method adopted for solving the bubble sort is called as odd- even transposition. Suppose an input series is B=(b1, b2, b3, b4......... bn) and each number is assigned to a particular processor. In the odd-even transposition,the sorting is executed with the help of two phases known as odd phase and even phase. In the odd phase, the elements stored in (p1, p2), (p3, p4), (p5, p6)......... (pn-1, pn) are compared according to the Figure and consequently exchanged if required i.e. if they are out of sort. In the even phase, the elements stored in (p2, p3), (p4, p5), (p6, p7)......... (pn-2, pn-1) are compared according to the Figure and consequently exchanged if required, i.e. if they are out of order. keep in mind, in the even phase the elements stored in p1 and pn are not exchanged and compared. The total number of phases needed for sorting the numbers is n i.e. n/2 odd phases and n/2 even phases. The algorithmic representation of the above discussed odd- even transposition is given below:

2084_Interconnection network in the form of a Linear Array.png

                                                Interconnection network in the form of a Linear Array


Related Discussions:- Sorting using interconnection networks

Configure ip settings for the linksys wrt300n- ccna, Configure IP Settings ...

Configure IP Settings for the Linksys WRT300N The best way to understand the following settings is to think of the WRT300N as being same to a Cisco IOS-based router with two di

Metrics for performance evaluation, Metrics For Performance Evaluation ...

Metrics For Performance Evaluation In this section, we would highlight different kinds of metrics involved for analyzing the performance of parallel algorithms for parallel co

What are intrusion detection systems, Question : a) Give three example...

Question : a) Give three examples of vulnerable services which are among the overwhelming majority of successful attacks. b) Name the five outside sources which can be the

Describe about the network equipment and connectivity, Describe about the N...

Describe about the Network equipment and connectivity Network equipment and connectivity should address various networking equipments such as routers, types of cables, as well

What is sliding window, What is Sliding Window? The sliding window is a...

What is Sliding Window? The sliding window is an abstract method that explains the range of sequence numbers that is the concern of the sender and receiver. In other words, he

How many types of twisted pair cable are there, How many types of twisted p...

How many types of twisted pair cable are there We can find two types of twisted pair cables, namely: Unshielded Twisted Pair Cable (UTP) and Shielded Twisted Pair Cable (STP).

Matrix computation and multiplication, In the following section, we shall d...

In the following section, we shall discuss the algorithms for solving the matrix multiplication difficulty using the parallel models. Matrix Multiplication Problem Let ther

Division of the address space, The IP class procedure does not split the 32...

The IP class procedure does not split the 32-bit address space into same size class and the classes do not have the same value of networks. A prefix of n bits gives 2n unique ne

Write Your Message!

Captcha
Free Assignment Quote

Assured A++ Grade

Get guaranteed satisfaction & time on delivery in every assignment order you paid with us! We ensure premium quality solution document along with free turntin report!

All rights reserved! Copyrights ©2019-2020 ExpertsMind IT Educational Pvt Ltd