Code+ course name

  • Category:
    Logic & Programming
  • Document type:
    Math Problem
  • Level:
  • Page:
  • Words:

6Computer Networks


Computer Networks

Question 1

  1. Such a network traffic issue can be solved by polling the router or interconnecting switch in some predetermined order. When polled, the router uses the full data rate of the Ethernet network to forward its backlog of packets to the internet or other networks. Between polls, routers accumulate packets in their queues; however, they do not forward or transmit until they are polled. The main disadvantage of the polling approach is that unsuccessful polls (polls in which packets are not present) increase the general overhead of the network. In addition, the receive operation latency increases because messages or packets are queued in the router up to the time that the polling event will occur[ CITATION Gho15 l 1033 ]. Because of the two disadvantages, polling is recommended for systems with heavy network load, where there is a low probability of unsuccessful polls and the receive latency can be lowered by the use of a specialized hardware or using a high polling period. One advantage of the polling approach is that it is not possible to over saturate a polled network with traffic[ CITATION Gho15 l 1033 ]. This is because the polling mechanism makes sure that the maximum traffic level is never exceeded. The excess network also does not affect the performance of the network. Another advantage is that polling makes the channel access to be predictable and fixed as access to the network is set at predetermined order.

  2. Token Bus networks used a coaxial cable to connect workstations to the main frame as shown in the figure below. The coaxial cable acted as the common communication bus and a Token Bus protocol created a token that was used to manage access to the bus. Only the station that had the token was allowed to transmit data. When it is done or a higher priority device wants to transmit, the station releases the token packet. The ensured that there was no collision within the network. When the traffic increases in a Token Bus, the data throughput rises to a certain level and then stabilizes[ CITATION Mou12 l 1033 ].

Code+ course name

Figure 1: A Token Bus Network

i) Collisions reduces the throughput on an Ethernet network using a bus topology, on the other hand, broadcasts improve throughput on the Ethernet network.

ii) In various levels of VLANs, switches, bridges, hubs, and repeaters can be used to influence collisions and broadcasts within an Ethernet network. Repeaters and hubs broadcast channels in a single collision domain. This means that there is a common collision domain and a collision can result whenever two nodes in a hub transmit simultaneously. A bridge is used to divide the network into two collision domains, which assist in reducing the congestion/collisions. Switches are like multiport bridges, they divide the network into different broadcast domains; hence, dramatically reducing congestion[ CITATION Ian16 l 1033 ].

Question 2

  1. The header is the only one checked by IPv4 for errors because the data that follows the header such as TCP, UDP, ICMP, etc. have their own checksums. It is necessary to check for errors even after frame CRCs because packet corruption can be caused by many errors apart from those introduced by the physical layer. Errors may be introduced by bugs in router and host software and hardware. Therefore, it is paramount that end-to-end checksums are used at the receiving host.

  2. Provided subnet mask:

Chosen IP address:

In binary numbers, we have: 11001001 01110101 01111011 01101001 11111111 11111111 11000000 00000000

We can add nine more 1’s to the subnet mask to make it look as follows:

11111111 11111111 11111111 11100000, which is in decimal.

Having extended the subnet mask to, we can create 8 subnets, each having up to 32 host addresses as follows:; Range to; Range to; Range to; Range to; Range to; Range to; Range to; Range to

  1. Fragmentation is whereby an IP packet that is too large for transmission is fragmented into smaller packets that it can be received by the destination system. There are two methods of fragmentation:

  1. IP router segmentation, in which the fragmentation is done in the routers

  2. IP path Maximum Transmission Unit (MTU) discovery, in which fragmentation is done by the sender (also supported by IPv6)[CITATION Lon13 p 75 l 1033 ].

Fragmentation may contribute to congestion as a result of dropping or discarding the packets. Discarded packets are normally retransmitted and a high retransmission rate may lead to congestion.

  1. The protocol will works as follows: The source host will transmit a packet to the destination host. The destination host will then check the packet for error. If a packet has an error, the destination host will respond to the source host with a negative acknowledgment frame. Finally, the source host will retransmit the packet.

  2. The figure below shows successful and unsuccessful delivery of datagrams from source host to destination host. The source host retransmitted packet 13 and received a negative acknowledgement (NACK (1)), which suggest that packet 11 had an error. The source then retransmit packet 11.

Code+ course name 1

Figure 2: Sequence Diagram

  1. Advantages: i) Only the corrupted packets are retransmitted; ii) It is not necessary for the destination host to buffer packets that have been received correctly.

Disadvantages: i) The destination host should be capable of reordering packets which are out of order; ii) The sender host has to buffer packets that were transmitted but have not been acknowledged

Question 3

  1. Decisions made by vector-distance routing protocols are said to be based on second-hand information because routers gets routing information from the neighbors connected directly to them, and the routers presume that their neighbors have also learned from their neighbors. These routing protocols presume that whenever a neighbor misses an update, it will learn from other neighbors or from next update[ CITATION Den14 l 1033 ]. For link-state protocols, their decision are based on first-hand information because any update information received by any router is propagated to all the routers and also stored in the link-state database.

Code+ course name 2

Figure 3: Provided Network Topology

Node A is designated as the current node and its state is (0, p)

Nodes C, E, F can be reached from node A and their distance values can be updated as follows:

dC = 0 + 1 = 1

dE = 0 + 4 = 4

dF = 0 + 1 = 1

From the smallest values, the status of node C and F are (1, p) and (1, p) respectively

Node E can be reached from node C and its distance value will be:

dE = 1 + 2 = 3; thus its status changes to (3, p)

Node B and D can be reached from node F and their distance values will be:

dB = 1 + 6 = 7

dD = 1 + 2 = 3; Node D has the smallest value and its status will be (3, p)

Node B can be reached from node E and its distance value will be:

dB = 3 + 1 = 4; Node D has the smallest value and its status will be (4, p)

Therefore the shortest path tree from A, will be as follows:

Code+ course name 3

Figure 4: Shortest-Path Tree

The final routing table at A will be as follows:

  1. i) The IPv4 header length specifies the size of the IP header, while the total length specifies the total length of the entire packet including the IP payload and IP header. The IPv6 payload length specifies the length of the IPv6 payload and the purpose of this field is to inform the routers the amount of information a specific packet have in its payload. The IPv4 total length field is similar to the IPv6 payload length except that the total length field includes the header whereas the payload length field specifies the length of the data contained after the header[ CITATION Cis14 l 1033 ].

  1. In IPv4, the protocol type field defines the transport protocol data such as TCP or UDP, whereas the Next Header field in IPv6 shows the new organization of IP packets. The IPv6 has the same structure and we can set the next protocol type to be TCP (6) or UDP (16), or it is possible to slot in Extension Headers between the TCP/UDP and IP payload[ CITATION Cis14 l 1033 ].

iii. The IPv4 time to live field is a theoretical value that was expressed in second and it is very difficult to estimate it. On the other hand, the IPv6 hop limit field specifies the number count of the maximum hops a packet can stay in the network before it is discarded or destroyed. In other words, it indicates the number of routers a packet can cross before it is discarded[ CITATION Cis14 l 1033 ].

Reference List

Cisco Networking Academy, 2014. Routing protocols companion guide. Indianapolis, IN: Cisco Press.

Deng, J., Wu, S. & Sun, K., 2014. Comparison of RIP, OSPF and EIGRP routing protocols based on OPNET. Available at:
[Online] [Accessed 26 August 2016].

Ghorbani, N. C., 2015. Review: Advantages and disadvantages of Mikrotik Nstreme protocols on wireless networks. International Journal of Computer Networks and Communications Security, 3(6), pp. 244-247.

Iannone, E., 2016. Telecommunication networks. 2nd ed. Boca Raton, FL: CRC Press.

Long, J., 2013. Storage networking protocol fundamentals. 2nd ed. Indianapolis, IN: Cisco Press.

Moussavi, M., 2012. Data communication and networking: A practical approach. 2nd ed. Boston: Cengage Learning.