Advantages And Disadvantages Of Network Topologies, OSI Model Vs TCP/IP Model, IPv4 Vs IPv6, POP3 States And DHT In P2P Networks

Advantages and Disadvantages of Star, Bus, and Mesh Physical Topologies

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

2. Explain why the OSI model is better than the TCP/IP model. Why hasn’t it taken over from the TCP/IP model?

3. Calculate the approximate bit rate and signal level(s) for a 3.5 MHz bandwidth system with a signal to noise ratio of 133.

4. Compare IPv4 and IPv6 private addressing. Discuss address ranges and relative sizes. Why don’t the same private addresses in different locations cause conflict on the Internet?

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

6. What is a Distributed Hash Table (DHT) and how is it used in P2P networks? Briefly, explain how a DHT works with an example of a P2P network.

1. Star topology is the connection of the nodes to the central hub. The advantages of this type of topology are many; the system is easy to install in a premise as the networking needs only a central hub and the wires that are used to connect the nodes or the computers to the hub (Bisht & Singh, 2015). The star network can be seen in most of the offices where the central LAN server can be observed with which the computers are connected.

The Bus topology is the networking method in which the central bus is used for connection between the nodes, the advantage of the system is lesser wire and a common backbone, in which transfer is decided by the bus master, which ensures connection between the nodes (Jiang, 2015). This topology is used in Industrial Ethernet where the RTUs send signals after equal interval of time

Mesh topology is the networking where each node is connected to each other through a separate communication media. This distributed networking makes the system most versatile for sensitive networks (Lim, 2016). This networking topology is used in MANET (Mobile Ad hoc Network) in which the mobile device is connected to multiple devices as it moves through the space, hence making a larger interconnected space.

Topologies	Advantages	Disadvantages
Star Topology	·         Node failure doesn’t affect the working of the system ·         Easy to install ·         Easy fault diagnosis ·         Updating or modification is easy	·         Expensive as it needs more cable ·         Failure of the central node will cripple the system ·         Terminators are required at the end of the cable ·         Problem identification is difficult
Bus Topology	·         Very reliable for small networks ·         Cheaper as requires lesser amount of cable ·         Easy to extend and update	·         Since only the bus master is given the power to transfer the next transfer has to wait till the transfer is complete ·         Bus failure will result in termination of services
Mesh Topology	·         This topology is most versatile as, in case of redundancy the data can be rerouted through other nodes ·         Provides best data privacy ·         Network errors are easier to diagnose	·         Costliest topology as the wire needed for the networking is very high

Table 1: The advantages and the disadvantages of Topologies
(Table Source: As created by the author)

2. The OSI (Open Systems Interconnection) model is developed by ISO (International Organization for Standardization) and aims to standardize the communication between the communicating devices to enhance the interoperability of devices, whereas TCP/IP is just a standard for interconnection. TCP/IP lacks the generic structure of the OSI model. OSI model consists of seven layers which divide the labor for interaction with one another on the contrary the TCP/IP model contains only five layers that makes the task of each layer complex. OSI layer has dedicated transport layer which ensures data packet delivery to the destination which saves the data loss which is common in TCP/IP. Finally the separate layer structure makes the OSI model much more versatile and easier to update compared to five layered TCP/IP model.

Differences Between the OSI Model and the TCP/IP Model and Why the OSI Model Failed

In his textbook author Tanenbaum (2003) have discussed in detail about the failure of OSI model. The failure of the model has been attributed to three major factors which are timing, technology and the implementation and politics.

The time was bad for implementation as the release of the model was delayed due to extensive research that has been carried out for the modeling and which resulted in extensive expenditure on TCP/IP. The technology was not up to the mark as few layers were empty and the others were overloaded (Severance, 2013). Due to these issues, the earlier models of the OSI were buggy. Finally, the patch up of TCP/IP with Unix buried the last hope of implementation on large-scale (Why is TCP/IP used rather than OSI? – 77624 – The Cisco Learning Network, 2016).

3.  The channel capacity:

       

4. Private addresses are the IP addresses that are not directly connected to the internet. These IP addresses are meant for an internal network like LAN, etc., the router which connects the computer to the internet uses tools like Network Address Translator (NAT) for connecting to the internet (Matousek Skacan & KoÃ…â„¢enek, 2013).

IPv4	IPv6
·         Address  length of IPv4 is 32 bits	·         IPv6 has 128-bit long address
·         They are represented  in decimal numbers	·         They are represented by the hexadecimal numbers
·         Two types of configuration are available (manual and automatic)	·         Only automatic configuration is available
·         Identification of packet flow is not available	·         Flow label is available forIPv6 header

Table 3: Comparison between IPv4 and IPv6
(Table source: As created by author)

IPv4 which has an address length of 32 bits has the maximum address of 10¹⁶-2, but the IPv6 which used hexadecimal digits instead of decimal has a wider range of 10²⁸ addresses. Hence IPv4 has an addressing constraint whereas IPv6 doesn’t have an addressing constraint.

The private addresses are used only on the local networks and are prohibited from using it in public internet. For the purpose of public interaction, the separate address is generated by NAT which removes the concept of IP conflict. Hence same IP address can be used in different networks without any conflict of interest.

5. The POP3 commands are highly dependent upon the present states which are namely the closed state, authorization state, and transaction and finally the update state after which the connection is closed again.

The authorization starts after the connection has been established between the client and the server. For the connection, the between the two sides TCP three-way handshake is used. During the authorization, the client sends the username and password, and the authorization is granted to the user. In the second state, the information about the transaction state is provided, and the data regarding the e-mails are provided (Fujiwara Newman & Yao, 2013). During the transaction, the STAT for the status, LIST for the content, RETR for returning the messages and DELE for deletion of the messages are the major signals. The messages are updated following the initial transactions, and apart from the above, various other signals are generated for the update. Finally, during the close session, QUIT command is generated for terminating the connection.

Comparison of IPv4 and IPv6 Private Addressing

Image 4: The four states of POP3 and how it moves through the stages
(Image Source: As created by author in Visio)

6. A hash table is the data structure used for implementation of the arrays for mapping the keys to the values which help in finding and retrieving the desired values. Hence the DHT could be understood as the distributed system that offers services similar to hash tables. P2P (Peer-to-Peer) is an application that is structured in a way to distribute the tasks among the peers so that the workload is distributed among the peers and the workload is reduced. In a P2P system, the every peer is an equal contributor and equipotent in term of resource allocation. The peers of the system allocate a portion of their computing resources for carrying out the tasks (He et al.,2016).

The most common type of structured P2P systems is implemented through DHT. In DHT, the hash tags are given to the various segments that are available with the different peers. The foundation of the DHT consists of abstract keyspace which are bit-strings. A program then slits the partnership of the keyspace among the peers, then an overlay network communicates between the nodes to which allows them to track the real file (D’Acunto et al., 2013). BitTorrent is a similar P2P program which uses its protocol for transferring and receiving the files. It is a two tier P2P that also allows searches across the network. BitTorrent is completely server less as the files are distributed all over the network and the distributed computing from the shared computers. The user of the service becomes a decentralized connected network.

References

Bisht, N., & Singh, S. (2015). Analytical Study if different Network Topologies.

D’Acunto, L., Chiluka, N., Vinkó, T., & Sips, H. (2013). BitTorrent-like P2P approaches for VoD: A comparative study. Computer Networks, 57(5), 1253-1276.

Fujiwara, K., Newman, C., & Yao, J. (2013). Post Office Protocol Version 3 (POP3) Support for UTF-8.

He, Q., Dong, Q., Zhao, B., Wang, Y., & Qiang, B. (2016). P2P Traffic Optimization based on Congestion Distance and DHT. Journal of Internet Services and Information Security (JISIS), 6(2), 53-69.

Jiang, R. (2015). A review of Network Topology.

Lim, F. P. (2016). A Review-Analysis of Network Topologies for Microenterprises. Small, 3, 15-000.

Matoušek, J., SkaÄÂan, M., & KoÃ…â„¢enek, J. (2013, April). Towards hardware architecture for memory efficient IPv4/IPv6 Lookup in 100 Gbps networks. InDesign and Diagnostics of Electronic Circuits & Systems (DDECS), 2013 IEEE 16th International Symposium on (pp. 108-111). IEEE.

Severance, C. (2013). Andrew Tanenbaum: Writing the Book on Networks.Computer, 46(12), 9-10.

Tanenbaum, A. S. (2003). Computer networks, 4-th edition. Ed: Prentice Hall.

Why is TCP/IP used rather than OSI? – 77624 – The Cisco Learning Network. (2016). Learningnetwork.cisco.com. Retrieved 19 September 2016, from https://learningnetwork.cisco.com/thread/77624