Effect Of Using Ip Version Computer Science Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Nowadays the world is witnessing a steady transition from IPv4 to IPv6 at the same time wireless has greatly emerged as the preferred technologies than wired for LANs because they provide cost effective way for connecting hosts into the network. The main aim of this paper is to evaluate the effect of using IPv4 and IPv6 over wired and wireless local area network (LAN) setups on the internet control message protocol (ICMP) based throughput of a LAN. Results are obtained by simulating wired and wireless LANs on Packet Tracer and a 22 experimental design technique has been employed for analysing results and drawing conclusions.

Keywords: Wired networks, Wireless networks, IPv4, IPv6, Internet Control Message Protocol (ICMP), Internet Protocol (IP), Transmission Control Protocol(TCP).

Introduction

Physical media and the protocols play a central role on the performance of a communication network. Both types of media wired and wireless have got their merits and demerits over one another when it comes to their usage for LANs. Wired medium provide much faster and secure connection within a network than the wireless. However, the wired infrastructure can't be used everywhere and wireless LANs can be built instantly anywhere. Data Link Control (DLC) layer of Open System Interconnection (OSI) model is divided into two sub layers Logical Link Control (LLC) and Medium Access Control (MAC). Focus has been made to these two sub layers in all the IEEE 802.3 (Ethernet), IEEE802.5 (Token Ring) and IEEE 802.11 (WLANs) standards [8]. In [9] it was indicated that changing the parameters like data rate, fragmentation threshold and RTS/CTS threshold influences the performance of WLANs. For ensuring communication level performance of a network and tackle burst data transmissions 100Mbps Ethernet is preferred [2]. Mobility of nodes also renders effect on the throughput in WLANs [15]. Network configuration plays an important role in determining the performance of a WLAN [1]. Presently the operating of IEEE 802.11 standard is still below the theoretical limit of throughput of the network [1]. The traffic load within a network has a direct impact on the performance of a network [5]. In [1] the authors have proposed various probabilistic models in order to study the performance of Wired and Wireless networks. Public Internet suffered a serious problem of the shortage of IPv4 address space which was overcome by the steady migration to IPv6 from IPv4 [6][10][11]. The migration is gradual or IPv6 and IPv4 were made to coexist together on the Internet [3][7]. Many problems which existed in the IPv4 were fixed in IPv6 such as the small address pool and others. The number of bits required to represent a single address in IPv6 is 128 which was earlier 32 bits only with IPv4.Improvements have been made to IPv4 in many areas such as network auto-configuration, routing and quality of service in IPv6 [4]. The performance of IPv4 and IPv6 on platforms such as Windows 2000, RedHat 7.3 and Solaris in terms of TCP/UDP throughput, CPU utilization, latency and web-based performance characteristics has been evaluated in [13] [14]. Furthermore in [12] performance of IPv4 and IPv6 has been evaluated on RedHat 9, Windows 2003 and FreeBSD 4.9 in terms of socket-creation time, TCP connection time and round-trip time. The main purpose of this paper is to determine the effect of choosing among IP protocol (IPv4 or IPv6) and connecting media on the Internet Control Message Protocol (ICMP) based throughput (i.e. number of eco packets that can be transmitted between two hosts in a same network per second). A 22 experiment design has been used in determining the impact of IP version and media selection on the throughput. The remainder of the paper is organized as follows. In Section 2 we describe the two Experiment setups each in turn divided into two stages, the results obtained from experiments are pronounced in Section 3. In Section 4 the experimentation results are analyzed by means of 22 experimental designs to aid the conclusions described in Section 5.

Experimental Setup

The experimentation is divided into four setups (2.1) employing IPv4 addressing scheme on a wired LAN setup and measuring ICMP throughput (2.2) employing IPv6 addressing scheme on a wired LAN setup and measuring ICMP throughput (2.3) employing IPv4 addressing scheme on a wireless LAN setup and measuring ICMP throughput and (2.4) employing IPv6 addressing scheme on a wireless LAN setup and measuring ICMP throughput. All LAN simulations will be carried out in a network simulator called packet tracer.

2.1 Ipv4 Addressing on a Wired LAN Setup 1

In this setup a wired LAN is being simulated. Four PC's and one server are connected with a 100 base-T layer 2 device named as CISCO 2950-24 switch as illustrated in Fig.1. Connections from PC's to switch are made by means of a straight cable Cat5 type cable. The PC's are allocated addresses 20.0.0.1-4 through either a dynamic host control protocol (DHCP) or static IP addressing mechanism. To measure throughput we send four ICMP packets along the network between hosts 20.0.0.1 and 20.0.0.2.

Fig. 1 wired LAN setup employing IPV4 addressing scheme

Fig. Output results when employing IPv4 on wired LAN

2.2 Ipv6 Addressing on a Wired LAN Setup 2

In this case we again connect four PC's and a server through a 100 base-T layer 2 device named as CISCO 2959-24 switch. Ethernet cables of Cat5 type are used as a media between hosts and the central switch. What is different from the experimental setup 1 is that, here we make use of IPv6 scheme of addressing in place of IPv4. The four PC's are allocated

addresses as 2340:1111:AAAA:2::1/64,2340:1111:AAAA:4::1/64, 2340:1111:AAAA:6::1/64,2340:1111:AAAA:8::1/64 respectively through auto configuration or static mechanism. Four ICMP packets are sent along the network between PC-3 and PC-3.

Fig. wired LAN setup employing IPv6 addressing scheme

Fig. 4 output results when employing IPv6 on wired LAN

2.3 Ipv4 Addressing on a Wireless LAN Setup 3

In this and the next setup experiments are conducted on wireless networks as shown in figure 5. Linksys simulate wireless network adapter is installed on each of the four PC's in order to connect them to a access point. PC's are allocated addresses 20.0.0.1-4 through either a DHCP or static mechanism. Once the network has been established four ICMP packets are sent along the network between hosts 20.0.0.1 and 20.0.0.2.

Fig. wireless LAN setup employing IPv4 addressing scheme

Fig. output results when employing IPv4 on wireless LAN

2.4 Ipv6 Addressing on a Wireless LAN Setup 4

In this case around again simulate a wireless LAN setup by connecting four PC'S with an Access Point. What is different from the Experimental Setup 3 is that in this case we shall be using ipv6 scheme of addressing in place of ipv4. The four PC'S are allotted addresses as 2340:1111:AAAA:2::1/64,2340:1111:AAAA:4::1/64,2340:1111:AAAA:6::1/64, 2340:1111:AAAA:8::1/64 respectively through auto configuration or static mechanism as illustrated in figure 7. After establishing the connectivity the four ICMP packets are sent along the network, the hosts 2340:1111: AAAA: 2::1/64 and 2340:1111:AAAA:6::1/64.

Fig.7 wireless LAN setup employing IPv6 addressing scheme

Fig. output results when employing IPv6 on wireless LAN

Results

Table 1: depicts the results got from the experimentation in Section 3.

Setup

Network Type

Addressing Schema

Transmission time

Throughput

1.

Wired

IPv4

15[m sec]

67

2.

Wired

IPv6

13 [m sec]

77

3.

Wireless

IPv4

21[m sec]

48

4.

Wireless

IPv6

18[ m sec]

56

Table Results table derived from Experimentation

When underlying network is Wired and Addressing Schema used is IPv4

Throughput=1000/15=67 ICMP packets per second.

When underlying network is Wired and Addressing Schema used is IPv6

Throughput=1000/13=77 ICMP packets per second.

When underlying network is Wireless and Addressing Schema used is IPv4

Throughput=1000/21=48 ICMP packets per second.

When underlying network is Wireless and Addressing Schema used is IPv6

Throughput=1000/18= 56 ICMP packets per second.

Analysis

A 22 experimental design a special case of a 2kfactorial design with k = 2is used. In this case, there are two factors each at two levels one being the physical infrastructure of the network (wired or wireless) and other being the IP addressing schema (IPv4 or IPv6) . Such a design can be easily analyzed using a regression model.

Consider the problem of studying the impact of choosing appropriate internet protocol and underlying network on the performance of a network. Two levels of each of these two factors are chosen for the simulation. The performance of the two Networks in terms of echo packets sent per seconds is listed in Table 2.

IP Addressing

Wireless

Wired

IPv4

48

67

IPv6

56

77

Table Performance in number of eco packet sent per second

Here define the two variables xA and xB as follows:

xA=

xB=

The performance y in terms of echo packets sent per second is regressed on xA and xB using a nonlinear regression model of the form as show in Eq.1

y= x0+ qAxA + qBxB+ qABxAxB (1)

In general, any 22 design can be analyzed as illustrated in the table 3

Experiments

A

B

Y

1

-1

-1

y1

2

1

-1

y2

3

-1

1

y3

4

1

1

y4

Table 3 Analysis of a 22 Design

The four observed responses are represented by y1, y2, y3 and y4 for fore setups respectively. Substituting the four observations in the model are given in Eq.2, Eq.3, Eq.4 and Eq.5 respectively

y1= q0 - qA- qB + qAB (2)

y2= q0+ qA - qB - qAB (3)

y3 = q0 - qA+ qB - qAB (4)

y4= q0+ qA + qB + qAB (5)

Solving these equations for qi's, one gets

q0=1/4{y1+ y2+ y3+ y4} (6)

qA =1/4{-y1+ y2- y3+ y4} (7)

qB=1/4{-y1- y2+ y3+ y4} (8) qAB=1/4{ y1- y2-y3+ y4} (9)

Expressions for qA, qB, and qAB are linear combinations of the responses such that the sum of the coefficients is zero. Such expressions are called contrasts. The coefficient of yi's in the equation for qA are identical to the levels of A, as listed in Table 3. qA can be obtained by multiplying the columns A and y in the table. This is also true for qB and qAB, both of which can be obtained by multiplying the respective level columns with the response column. This observation leads us to the sign table method for calculating effects, as illustrated in the table 4.

I

A

B

AB

Y

1

-1

-1

1

48

1

-1

1

-1

56

1

1

-1

-1

67

1

1

1

1

77

248

40

18

2.0

Total

62

10

4.5

0.5

Total/4

Table 4 Sign Table Method of Calculating Effects in a 22 Design

The regression equation becomes:

y = 62+ 10xA+ 4.5xB+0.5xAxB

The result is interpreted as follows. The mean performance is 62 packets sent per sec; the effect of IP version on throughput is 4.5 packets per second; the effect of network type on throughput is 10 packets per second; and the interaction between IP version and underlying network is 0.5 packets per second.

4.1. Variation Allocation

Total variation of y or Sum of Squares Total (SST) =4qA 2+ 4qB 2+4qAB 2

=4*100+4*20.25+4*0.25

=482

Sum of Squares due to A (SSA) =4qA 2= 400

Sum of Squares due to B (SSB) =4qB 2 = 81

Sum of Squares due to AB (SSAB) =4qAB 2= 1

Fraction of Variation due to A (IP version) =SSA/SST= 400/482=82.9%

Fraction of Variation due to B (Network type) =SSB/SST= 81/482=16.80%

Fraction of Variation due to AB (Interaction between IP version and Network type)

=SSAB/SST= 1/482=0.2%

Relative Importance of factors = A (Network type)> B(IP version)> AB(Interaction between IP version and Network type)

Conclusion

In this paper experimentations are carried out to evaluate the effects of physical infrastructure and IP addressing schema on the performance of the LAN in terms of ICMP based throughput. A 22 experimental design has been employed for the same. From the analysis carried out in Section IV using the 22 design following could be concluded: The mean performance using both IPv4 and IPv6 has increased by 7.8% from the previous case when we were employing IPv4 only for both wired and wireless networks. When we employed IPv6 for networks (both wired and wireless) it is found that throughput increases by 15.6% from the stage when only IPv4 is being used. Effect on throughput by virtue of selecting an appropriate network infrastructure (wired or wireless) is more than two times than the effect of selecting the IP version. Mutual interaction among the appropriate IP version & network infrastructure is least as compared to all other independent interactions. Finally it has been found that employing IPv6 in place of IPv4 has noticeably increased the performance of both type of networks infrastructure but the experimentation proved that IPv6 works better for wireless networks than for wired ones.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.