Routing Concepts, TCP Congestion Control Techniques, And Essay.

1.From the point of view of router R4, what is the next-hop address for a packet addressed to host 161.22.0.15/18?

2.From the point of view of router R1, which of its interfaces would it choose for a packet being sent to network 161.22.0.0/18?

3.A host with an IP address of 200.11.60.36/24 has just sent a packet to a host with address 150.32.0.240/18.   How many hops is required between source and destination?

4.A packet originating from network 220.10.40.0/24 arrives at router R1, however, R1 determines that the destination network is not in its routing table.  What does R1 do with the packet?   

5.A packet arrives at router R2 with a destination address of 140.21.0.10/22.  Which interface port does R2 forward the packet out of?

6.A packet at router R3 has a destination address of 220.10.40.5/26.   What next-hop address would R3 use for this packet?   

7.A packet is waiting at router R4 for forwarding.  If the next-hop was a “direct delivery”, which of these three networks is the destination network?  150.3.0.0/16, or 150.32.0.0/18, or 220.10.40.0/24?

An IP datagram 5,400 bytes long with no options arrives at a router, which determines that the next destination has an MTU of 1,500 bytes.  Use the Answer Template to complete the following questions, showing your calculations and reasoning.

a)Assuming that the router decides to fragment the packet into 4 fragments, determine a correct size for each fragment, and identify the starting byte and ending byte of each fragment .

b)Calculate the fragmentation offset for each fragment .

c)State whether the total number of bytes from all 4 fragments leaving the router will be greater than the initial datagram size that arrived, or less than the initial datagram size, and the reason .

Exercise Questions

This assessment task requires you to demonstrate your knowledge of routing concepts by completing a number of exercise questions.  

The questions are designed to help you to achieve the unit learning outcomes as listed in the unit profile.

You must do this assignment on your own – it is not a group assignment.

These questions will require more time and effort than the first assignment so plan ahead and start as early as possible. Question #3 may require additional research and analysisto complete.

Type all your answers in the ‘Template for Your Answers’ Section of this document and upload only that template. You can do that by copying the Template section into a new Word document for uploading.  Answers that are not typed into the “Template for Your Answers” section may not be marked, or may be returned to you for re-typing and re-submission – late penalties will apply.

Where instructed, you must show the steps you took to arrive at your answers. Write your answers in your own words to avoid potential plagiarism and copyright violations.

You must submit the Answer section as a Word file (.doc or .docx).  Do not submit PDF’s or any other type of file without express permission from the Unit Coordinator.

Plagiarism Procedures can be found in the CQUniversity Policies section of the Unit Profile.

Assessment Requirements and Marking Criteria

There are 3 main questions each with sub-questions and the requirements are stated for each one.  You must answer all questions and their sub-questions.   Marks are indicated in the Answer Template.

The questions will be marked on correctness,logic and clarity, and addressing all parts of the question. 

The Assignment Questions begin on the next page.

Given the following network diagram, assume that all the networks shown are aware of each other and have fully updated routing tables.  Answer the questions that follow.

1.From the point of view of router R4, what is the next-hop address for a packet addressed to host 161.22.0.15/18?

2.From the point of view of router R1, which of its interfaces would it choose for a packet being sent to network 161.22.0.0/18?

3.A host with an IP address of 200.11.60.36/24 has just sent a packet to a host with address 150.32.0.240/18.   How many hops is required between source and destination?

4.A packet originating from network 220.10.40.0/24 arrives at router R1, however, R1 determines that the destination network is not in its routing table.  What does R1 do with the packet?

Network Diagram

5.A packet arrives at router R2 with a destination address of 140.21.0.10/22.  Which interface port does R2 forward the packet out of?

6.A packet at router R3 has a destination address of 220.10.40.5/26.   What next-hop address would R3 use for this packet?

7.A packet is waiting at router R4 for forwarding.  If the next-hop was a “direct delivery”, which of these three networksis the destination network?  150.3.0.0/16, or 150.32.0.0/18, or 220.10.40.0/24?

8.Complete the information in the routing table for router R2 as shown in the Answer Template for networks 150.3.0.0/16, 150.32.0.0/18, and the Default network.  Show the masks in longest mask order using CIDR format).

1.Write a brief summary of the congestion controls currently available in TCP as covered in this Unit

2.Identify and explain two problems with current congestion controls in TCP that are pointed out in the articles

3.Summarize in your own words the difference(s) between the current TCP congestion controls and Google’s new BBR protocol

4.The Network World article points out that it is difficult to get a new protocol accepted as a global standard for TCP/IP.  Why do you think this is the case?   Give carefully thought out reasons for your answer. 

Important:  for every direct quotation you use from these two sources or any other source, you must immediately, after the quote, provideyourownexplanation of the quotation (for example, explain why are you quoting it, how does it help answer the question, how does it support what you are saying?) – marks will be deducted for failure to do so.  In addition, correct referencing conventions must be used throughout your work using the Harvard referencing convention.  Your answers will be marked on clarity, logic, relevance, use of own words and fully addressing all parts of each question.

Remember that quotations alone will not be accepted as your explanation of the questions.  Quotations can support your explanations, but you must still provide the explanations yourself.   Best way forward is to keep direct quotations to a minimum, and use your own words.

TCP Reno

This congestion control technique implements the fast recovery algorithm where the sender retransmits packets when it receives duplicate acknowledgments before time out. This makes for faster retransmission. It invokes the congestion avoidance algorithm without the need of going to start slow mode to maintain reasonably consistent network performance (Abed, Ismail, and Jumari, 2012).

TCP BBR

This is a model-based congestion control technique whose behaviour is dictated by the network path that packet traffic flows through. It is based on two basic parameters, the Bottleneck Bandwidth and the Round trip propagation time (BBR) and determines the fastest way in which data can be sent in the different routes (networkworld.com,2017).

TCP Congestion Control Techniques

Quick UDP connections (QUIC)

This is a congestion control technique developed by Google for low-latency Internet transportation protocol over UDP. It utilises bandwidth estimation to avoid congestion

This is a loss based congestion control technique which reduces window size when packet loss occurs. The window size is a linear function of time.

CUBIC TCP

This is an improvement on BIC but less aggressive and TCP friendly. The window size is a cubic function of time (Wang, Wen, Han, Zhang, J., Li, and Xiong, 2013).

Compound TCP

This is a delay based and loss based congestion control technique developed by Microsoft Corporation to achieve bandwidth scalability and

TCP fairness (reduction of sending rate when congestion is detected).

Shallow buffers-In loss based congestion control, when switches with shallow buffers are used, packet loss occurs before congestion. This triggers a multiplicative reduction of sent packets to unnecessarily low levels

Deep buffers- In loss based congestion control, when bottleneck links with deep buffers are used, delays occur because the congestion control tries to fill up the buffers (bufferfloat).

The BBR technique is proactive in its behaviour in that it first assesses the network path and estimates the required bandwidth and the round trip propagation time to determine the best route to use to send data. In contrast, the TCP control methods were designed to slow down data traffic once a congestion is detected. This makes BBR faster and more reliable.

One of the main reasons why it is difficult to get a new protocol accepted as a global standard for TCP/IP is because of the standardization issue. The protocol has to be in line with the relevant international standards bodies such as ISO and RFC for global implementation of the new standard.

Another reason is that there are politics always involved in the development and acceptance of these protocols. According to Larsen (2012) both technical and political dynamics have shaped the TCP/IP protocol since the 1970s. Within that span of time, many protocols have been rejected by internet developers.

References

Abed, G.A., Ismail, M. and Jumari, K., 2012. Exploration and evaluation of traditional TCP congestion control techniques. Journal of King Saud University-Computer and Information Sciences, 24(2), pp.145-155.

Cardwell, N., Cheng, Y., Yegane, S.H. & Jacobson, V., 2017, BBR congestion control, Internet Congestion Control Research Group

Larsen, R., 2012. The Political Nature of TCP/IP. Momentum, 1(1), p.20.

Wang, J., Wen, J., Han, Y., Zhang, J., Li, C. and Xiong, Z., 2013. CUBIC-FIT: a high performance and TCP CUBIC friendly congestion control algorithm. IEEE Communications Letters, 17(8), pp.1664-1667.