Advantages And Disadvantages Of Star, Bus, And Mesh Physical Topologies And TCP/IP Protocol Suite Essay.

Question 1: Discuss the advantages and disadvantages of star, bus, and mesh physical topologies. Provide real examples of each type.

Question 2: Explain encapsulation and decapsulation in a five layer TCP/IP protocol suite. How does multiplexing and de-multiplexing differ from encapsulation and decapsulation? 

Question 3: Calculate the approximate bit rate and signal level(s) for a 6.8 MHz bandwidth system with a signal to noise ratio of 132.

Question 4: Explain why the OSI model is better than the TCP/IP model. Why hasn't it taken over from the TCP/IP model? Discuss the advantages and disadvantages of both models. 

Question 5: What is the total delay (latency) for a frame of size 5 million bits that is being sent on a link with 10 routers each having a queuing time of 3.5 µs and a processing time of 1.8 µs. The length of the link is 1900 km, the speed of light inside the link is 2.2 x 108 m/s, the link has a bandwidth of 8 Mbps. Which component(s) of the total delay is/are dominant? Which one(s) is/are negligible?

Question 6: According to RFC1939, a POP3 session is one of the following states: closed, authorization, transaction or update. Draw a diagram and explain to show these four states and how POP3 moves between them.

Star Topology

A topology in computer networking refers to a specific way of organizing network elements such as computers, printers and networking hardware (Nizam, 2014). Mitchell (2017) defines a network topology as the physical arrangement of network devices in a computer network. There are many network configurations including star, bus, partial mesh, full mesh, ring and tree topologies (Mitchell, 2017a).  Star topology is arranged in such a way that it has a central node such as a hub or switch where all other devices connect to (Meador, 2008). All traffic in a star topology originates from the central device in the configuration.  The central node usually is a fast, independent computer that is responsible for channeling all data traffic to all other connected devices (Telecom ABC, 2016). The central device therefore controls and manages all the devices that have been attached to it.  A sample illustration of a star topology is shown in figure 1 below.

Figure 1: Simple star topology (Source: Sparrow 2015)

Star topology networks are mostly implemented in small computer networks such as Local Area Networks (LANs) both in office and home networks

• All devices have access to centralized network resources
• Star topology is simple to troubleshoot
• If a device cable malfunctions , it can’t affect the others
• Easy to extend a star topology by adding  extra nodes
• Low costs and easy to maintain

• The entire network shuts down if the central node malfunctions
• Requires a lot of cable to setup in case of many devices
• If the central node is a hub, data is forwarded to all devices which causes congestions and slows down the network

Bus topology is an arrangement such that all network devices are connected together by straight line cable as shown in the figure below (Meador, 2008). When a network device wants to interconnect with another device through the Internet, it sends  a  message on the shared cable so that all devices can be able to see. However, only the required intended recipient essentially takes and works on the message (Mitchell, 2017b).  All the transmitted data goes to devices in the network but the receiving node is usually the destination address.  A simple illustration of a bus topology network is shown in figure 2 below.

Figure 2: Simple bus topology (Source: Sparrow 2015)

Real life application

Bus topology network are better applied in small work offices and school computer laboratories.

• Bus topology is easy to set up
• Easy to lengthen the connecting cable and extend the network
• Useful for small networks
• Compared to star topology, bus topology uses less cable
• Does not require the use of any dedicated network hardware

• Hard and complicated to troubleshoot
• Network disturbance when computers are removed or added
• If the connecting cable malfunctions, the entire network shuts down
• Bus topology suffers signal degradation if the connecting is too long
• Not usable for large networks
• Less security since all devices transmit on same cable
• Requires terminations at the cable ends
• Network turns out to be slow when many devices are connected as only a single cable is used for data transmission.

In Mesh topology configuration, every device is coupled to each other as shown in the illustration below.  Mesh topology network can either be partial where devices are not necessarily connected to all other devices or full topology where all devices are connected to one another. The topology creates several routes such that every device is connected to all the other devices (full mesh) hence presents the idea of routes (Mitchell, 2017c). The devices are so connected in order to provide fault tolerant and optimal data transfer (Nizam, 2014). A partial and full mesh network topologies has been shown in figure 3 below.

Bus Topology

Figure 3: Mesh topology network (Source: Concept Draw)                                                 

Real life application

• Mesh topology networks are usually and performs very well in social networking applications where devices communicate with one another in real time mode.
• Useful for wireless technologies including ad-hoc networks (Mitchell, 2016)

• Mesh topology can handle a lot of network traffic because it has more cable routes
• Smooth data transmission
• Easy to add extra network devices
• When one cable fails, traffic can be sent using alternate routes
• Offers redundancy and reliability due to alternate routes  
• Provides security and privacy

• Requires a lot of cable hence expensive
• Difficult to manage for large networks
• Requires experienced administrators due to numerous ports and hence expensive to maintain

Transmission Control Protocol (TCP)/ Internet Protocol (IP) protocol suite provides a set of protocols prearranged in different layers used Internetworking today (Tanenbaum, 2003). TCP/IP protocol suite aims at linking many networks smoothly.  (Tanenbaum, 2003a). IP designates transferring data packets to the correct path while TCP aims involves sending data packets without the occurrence of data loss or data misrepresentation. (Burke, 2014). As shown below in figure 4, TCP/IP protocol suite has four layers.

Figure 4: TCP/IP Protocol Suite Layers (Source: Tanenbaum 2003)

Encapsulation and Decapsulation in TCP/IP Protocol Suite

In a layered architecture, data is packed at very layer depending on its function. The process of wrapping up data on every layer is referred to as data encapsulation (Thomas, 2015).  A bundle of important data is added to the real actual data of every layer when data travels from the top most layer to the bottom layer of the TCP/IP protocol suite.   The data packet comprising of data and header information from the top layer usually becomes the information that will be wrapped up again in the next packaged lower layer and the layer header will be added.  The header data is the supplementary information positioned at the start of a data packet when it is being transferred. On the receiver side, information is used to excerpt the sent data from the compressed data packet. TCP/IP encapsulation is shown in figure 5 below.

Figure 5: TCP/IP Encapsulation

Decapsulation is the opposite of encapsulation and transpires when a data packet is acknowledged on the receiver side. Received data travels up from the bottom layers up to the top layers of TCP/IP protocol suite where unpacking of the header data happens and the header information used to send the packet to the next layer. This happens until a data packet is in its original form in the upper most layer for the sender to read.  TCP/IP de-capsulation process is shown in the figure 6 below.

Figure 6: TCP/IP Decapsulation

Multiplexing and De-multiplexing

Multiplexing refers to the process where many data signals from different sources are combined together and sent over a communication channel at the same time (Burke, 2015a). On the receiver side however, the separated data signals are combined back together, usually referred to as de- multiplexing (Burke, 2015b).  Encapsulation deals with adding layer functional data to a data packet while multiplexing refers to integrating multiple data packets from different sources for transmission and vice versa. As the TCP/IP protocol suite makes use of protocols at particular layers, multiplexing occurs at the source while de-multiplexing happens on the destination side. In such a case multiplexing means that a protocol may encapsulate a data packet at a particular layer from other various layer protocol; de-multiplexing on the other hand, means that a given protocol may de-capsulate and transport a data packet to other layer protocols. Figure 7 below shows multiplexing and de-multiplexing at the source and destination.

Mesh Topology

Figure 7: Multiplexing and De-multiplexing (Source: Forouzan 2013)

To calculate bit rate I use Shannon’s formula as shown below  

Maximum capacity = Bandwidth * log₂ (1 +SNR)

Bandwidth = channel bandwidth, SNR = Signal Noise Ratio, Maximum capacity = measurement of channel in bits per second (bps). Applying Shannon’s formula above,

Max channel capacity = 6.8 MHz log ₂   (1+132)

          6.8 * log ₂ (133) = 6.8 * 7.055

                = 47.6 Mbps

Therefore the bit rate = 47.6 Mbps

To calculate the number of signal levels, apply the formula below

Bit rate = Bandwidth * 1og₂ N *2

Bandwidth = bandwidth of the channel, N = signal levels used for data representation, Bit rate = bits per second (bps).

47.6 Mbps = 2 * 6.8 MHz * log ₂ L

47.6 Mbps = 13.6 * log ₂ L

Log ₂ L = 47.6 / 13.6

Log₂ L = 3.529

L = 2 ^3.5

L = 11

No of signal levels for a 6.8 MHz bandwidth system = 11

 Open Systems Interconnection (OSI) model was presented in the 1900s to cover all facets of network devices communication (Mitchell, 2017d). The purpose of the OSI model is to enable communication between different systems without the need to change the underlying hardware and software. It comprises of a set of protocols that allow diverse systems to communicate irrespective of their structure design (Burke, 2016).  OSI represents a model for designing robust, flexible and interoperable network architectures. The OSI model is comprised of layers used for designing computer    network systems and permits communication among all kinds of systems. It comprises of seven layers and each layer provides a step in the development of sending data packets in a network

TCP/IP model has four layers including Data-Link, Internet, Transport and Application. Each of these layers in the TCP/IP model parallels to a layer  or more of the Open Systems Interconnection (OSI) model (Microsoft, 2017).

• TCP/IP comprises of four layers while OSI consists seven layers as shown in figure 8 below
• The TCP/IP model does not link to the OSI model since it  was created way  before the OSI model
• TCP/IP was only designed to resolve a particular set of issues and not to function as an all-purpose model for all network communication

Figure 8: OSI Vs TCP/IP Model (Source: Tanenbaum 2003)

• Why the OSI model is better than the TCP/IP model
• OSI supports both connectionless and connection-oriented
• OSI is a protocol independent standard while TCP/IP is protocol reliant Why OSI has not taken over from the TCP/IP model
• Some layers and protocols in the OSI model are not yet  fully well-defined
• OSI model did not display a show enough performance to enable  the Internet authorities  to change from the TCP/IP model to the OSI model Advantages of OSI model of include the following
• OSI is a generic and standard model used in computer networking (Chaudhari 2016)
• Flexibility with all networking protocols
• Changes in one layer do not affect other layers
• It permits communication between different types of software and hardware
• OSI Model offers support for connection-less as well as connection-oriented services
• Easy to troubleshoot systems following the OSI model structure
• OSI model offers security and adaptability as all services are bundled in a single layerAdvantages of TCP/IP model
• It provides less rules as long as all the general guidelines are accomplished
• Establishes connection between different kinds of computer networks
• OSI model operates independent of the computer operating system.
• Provides internetworking among different organizations.
• Provides a  scalable, client-server architecture

Data size is 5 million bits = 5 * 10 ⁶

Processing time = 1. 8 * 10 ⁶

Queuing time = 3.5 * 10 ⁶

Link length (distance in km) = 1900 km * 10 ³

Speed of light in the link = 2.2 * 10⁸m/s

Bandwidth of the link = 8Mps * 10 ⁶

To calculate delay (latency) apply the following formula

Propagation time = distance/ propagation speed (1900km * 10 ³ / 2.2 * 10 ⁸ m/s) = 0.0086m/s

Encapsulation and Decapsulation in TCP/IP Protocol Suite

Transmission time = data size / Bandwidth (5 * 10 ⁶ / 8 * 10 ⁶) = 0.625m/s

Latency (delay) = transmission time + propagation time + queuing time + processing time

Delay = (0.625 + 0.0086 + 3.5 * 10 ⁶ +   1.8 * 10 ⁶)

The Post Office Protocol (POP) provides users with a platform to access and retrieve mails. (Microsoft, 2003). Post Office Protocol allows users to access their emails without online connection by allowing them to receive mail from servers to client machines.  POP3 is the recent version of Post Office Protocol (POP).  A POP3 session traverses through several states including authorization, update and closed when it is active (Microsoft, 2003).  The POP 3 session starts when a client computer establishes a connection to a server. The server directs a hello greeting and comes into the authorization state. The client then categorizes itself for a connection to the POP3 server. If the verification is successful, the period changes to a transaction state where the client computer performs mail procedures. After being done with all mail procedures, the client computer finishes the session and the session state moves to update and the server discharges resources attained in the course of transactions and closes the TCP connection changing it to a closed state. Figure 9 below shows POP3 states from how they follow each other from when there is no connection to the update session.

Figure 9: POP3 Session States

References

Burke, J. TechTarget: What is multiplexing? Retrieved from https://searchnetworking.techtarget.com/definition/multiplexing

Concept Draw (2017). Network Topologies .Retrieved from https://www.conceptdraw.com/How-To-Guide/network-topologies

Chaudhari, A. (2016). 12 Advantages and Disadvantages of OSI model Layered Architecture. Retrieved from https://www.csestack.org/advantages-disadvantages-of-osi-model-layered-architecture/

Forouzan, B, A. (2013). Data Communications and Networking. (5^th Ed.). McGraw-HilI

Mitchell, B. (2016). What Is a Mesh Network? Retrieved from https://www.lifewire.com/what-is-a-mesh-network-817885

Microsoft (2017). TechNet: TCP/IP Protocol Architecture. Retrieved from https://technet.microsoft.com/en-us/library/cc958821.aspx

Microsoft (2003).TechNet: How POP3 Service Works. Retrieved from https://technet.microsoft.com/en-us/library/cc737236(v=ws.10).aspx

Meador, B. (2008). A survey of computer network topology and analysis examples. Retrieved from https://www.cs.wustl.edu/~jain/cse567-08/ftp/topology/index.html

Mitchell, B. (2017). Lifewire: Introduction to Computer Network Topology.  Retrieved from https://www.lifewire.com/computer-network-topology-817884

Nizam, A. (2014). Networking Basics: Advantages and Disadvantages of Using Mesh Topology. Retrieved from https://www. /Advantages%20and%20Disadvantages%20of%20Using%20Mesh%20Topology.html

Sparrow, P. (2015). What is Bus topology? Bus topology: Advantages and disadvantages. Retrieved from https://www.ianswer4u.com/2011/05/bus-topology-advantages-and.html#comment-form

Sparrow, P. (2015), Star topology: Advantages and disadvantages. Retrieved from https://www.ianswer4u.com/2011/05/star-topology-advantages-and.html#comment-form

Tanenbaum, A. (2003) .Computer Networks. (4^th Ed.). Prentice Hall

Telecom ABC (2016). Star topology. Retrieved from https://www.telecomabc.com/s/star.html

Thomas, J. (2015). TCP/IP Data Encapsulation and Decapsulation. Retrieved from https://www.omnisecu.com/tcpip/tcpip-encapsulation-decapsulation.php