Backbone Networks

Part 1: Backbone Networks

Review the diagram of the backbone network below, and then answer the series of questions.

1. The textbook describes three types of backbone networks – switched, routed, and a mixture of the two. What type of backbone network does this diagram illustrate? Explain your answer.

The backbone architecture refers to the way in which the backbone interconnects the networks attached to it and how it manages the way in which packets from one network move through the backbone to other networks. These architectures are routed backbones (routers that move packets on the basis of network layer addresses), switched backbones (switches that move packets based on data link layer addresses), and virtual LANs (switches that move packets through LANs that are built virtually, not using physical location) [1].

The above diagram illustrates a switched backbone network because all the LAN segments are connected with layer-2 switches and there is no routing involved to move packets from one Local Area Network to another.

2. The textbook also describes that three layers exist in a backbone network. For each labeled computer and device in the diagram above:

a. indicate which of these layers the computer or device is a part of.

Computer Y: Access layer

Switch A: Access layer

Switch B: Access layer

Switch C: Distribution

Switch G: Core layer

Computer Z: Access layer

Switch D: Access layer

Switch E: Access layer

Switch F: Distribution layer

Switch H: Core layer

b. explain the role the computer or device plays in the backbone network.

Computer Y: It is a host (node) and it sends as well as receives data over the network.

Switch A: Switch A is at the access layer and it provides end-connectivity of hosts in LAN1 with the rest of the network.

Switch B: Switch B is at the access layer and it provides end-connectivity of hosts in LAN2 with the rest of the network.

Switch C: Switch C is at the distribution layer and it interconnects the hosts of LAN1 and LAN2 with each other. It also connects with switch G (at the Core layer) to provide connectivity with the devices on LAN 3 and LAN 4 (Building 2).

Switch G: Switch G is at the Core layer and it provides backbone connectivity between Building 1 and Building 2. On one side it connects with Switch C in building 1 and Switch H on the other side.

Computer Z: It is a host (node) and it sends as well as receives data over the network.

Switch D: Switch D is at the access layer and it provides end-connectivity of hosts in LAN3 with the rest of the network.

Switch E: Switch E is at the access layer and it provides end-connectivity of hosts in LAN4 with the rest of the network.

Switch F: Switch F is at the distribution layer and it interconnects the hosts of LAN3 and LAN4 with each other. It also connects with switch H (at the Core layer) to provide connectivity with the devices on LAN 1 and LAN 2 (Building 1).

Switch H: Switch H is at the Core layer and it provides backbone connectivity between Building 1 and Building 2. On one side it connects with Switch F in building 2 and Switch G on the other side.

3. If Computer Y were to broadcast a data-link layer frame intended for all computers, the computers on which LANs would receive that broadcast? Explain your answer.

All the computers in all the LANs would receive the broadcast. Recall that switches do not break broadcast domains. As it is a fully switched network, no VLANs exist and there is no layer-3 (routing) involved. All the LANs depicted in the diagram comprise a single broadcast domain. Therefore, all the devices would receive the broadcast.

4. If Computer Z intended to send a message specifically to Computer Y, would it be able to address the data-link layer frame with Computer Y’s MAC address? Explain your answer in detail.

Yes, it would be able to address the data-link layer frame with Computer Y’s MAC address. As all the LANs (including all the switches and all the computers) are in a single broadcast domain, the ARP request made by computer Z would be forwarded through switches D, F and H, then Switches G, C and A respectively. Then it would reach Computer Y which would respond by giving its MAC address and the ARP reply would travel in the opposite direction from Switch A, C, G, H, F, D and finally reach back to Computer Z. Then Computer Z would use this MAC address as the destination MAC address in the data-link layer frame to send a message specifically to Computer Y.

5. Now imagine that Switch G and Switch H were replaced with layer-3 routers. With that in mind, answer the sub-questions below.

a. If Computer Y were to broadcast a data-link layer frame intended for all computers, the computers on which LANs would receive that broadcast? Explain your answer in detail.

b. If a layer-3 router is used in place of switch G and a broadcast data-link layer frame is sent by Computer Y, only the computers in LAN 1 and LAN 2 would receive the broadcast because when the traffic will reach at layer-3 router G it would discard broadcast traffic. Recall that a router is a unicast device and it cannot forward broadcast traffic in any situation.

If Computer Z intended to send a message specifically to Computer Y, would it be able to address the data-link layer frame with Computer Y’s MAC address? Explain your answer in detail.

No, Computer Z would not be able to address the data-link layer frame with Computer Y’s MAC address if it intended to send a message specifically to Computer Y because replacing the Switches G and H would break the single broadcast domain into two separate broadcast domains. MAC addresses are only used for host-to-host delivery within a specific IP Network (LAN). Router H would act as a gateway for Computer Z. When Computer Z would send an ARP request to find the MAC address of Computer Y, the ARP request would be replied by the Router H as ARP request cannot be forwarded from one IP Network to another IP network (Building 1 is a separate IP Network and Building 2 is another IP Network). So Router H would reply the ARP request by sending the MAC address of its interface to which Switch F is connected. Computer Z would use this MAC address as the destination MAC address and forward this data-link layer frame to Router H which would then forward the traffic.

c. Would this new configuration offer better performance than the original configuration? Explain your answer in detail.

Yes, this new configuration would definitely enhance performance by many folds. Previously the Network was a flat, Switched Network and contained only a single broadcast domain. Each time a device on that flat Network made a broadcast request, all the devices including computers and switches would be affected by that broadcast storm. This is a huge disadvantage of a flat network having no routing involved. Using layer-3 routers breaks down the single broadcast into two separate broadcast domains so that all the broadcast traffic within a domain is confined only to that domain. This provides a huge advantage and offers much better performance as compared to the traditional flat switched network.

Another very promising feature of using layer-3 routers is that we can use Virtual Local Area Networks (VLANs) to further break down the broadcast domain into even smaller domains. Implementing VLANs provide additional security and even better performance. Therefore, this new configuration is very efficient as compared to the previous one.

d. Would this new configuration offer other advantages and disadvantages over the original configuration? Explain your answer in detail.

Advantages of Routed Networks

Routed Networks are more powerful, efficient, secure and easier to manage because they use a hierarchical approach rather than a flat design. Routed Networks implement routing protocols which also know the best paths to other networks (including the Internet) typically without additional configuration. Routing provides the backbone for very wide networks because routers regenerate data indefinitely across networks with no degradation [2]. When the network size increases, a routed network can be better since we will have better control about where people are sitting. It is easier to keep a structure in the network. Moreover, we can add some security on the routers and have a better control over different flows. When you have a flat network, everything would fall into one broadcast domain, which would mean the broadcast would be seen by all. There are certain broadcasts that we just wouldn't want to block. ARP is a case in point. Without ARP you would have to manually enter every MAC address on every device which is not practical with a large number of hosts [3]. Same is the case with DHCP.

Disadvantages of Routed Networks

Routers require skilled network managers in order to configure, deploy and manage. The more routers there are in a particular network, the more planning needs to be done. Routers are also much more expensive than switches because of this. Routers have a higher networking overhead cost because they're much more powerful than lower level networking hardware. For this reason, planning a network should include only as many routers as necessary.

Part 2: Wide Area Networks (WANs)
Review the architectural diagram of the WAN below, which illustrates a single organization’s WAN links between its campuses across the United States.

6. The textbook describes four types of WAN configurations – ring, star, full mesh, and partial mesh. Which configuration is illustrated in the diagram above? Explain your answer.

A star topology involves a central location serving as the hub in the design, with sites branching off the hub like spokes on a wagon wheel [4]. In a star topology, a failure to a single site (other than hub) will not affect the other sites on the WAN. Due to the importance placed on the central location serving as the hub, this site will benefit from redundant routers. As seen in the above diagram, all the sites are connected through a central site (i.e. Kansas City) which is serving as the hub and all other sites as spokes.

Therefore, this configuration is a star WAN topology.

7. Explain how a message sent from the Portland campus would travel to successfully arrive at the Miami campus.

Some type of WAN technology should be deployed like Frame Relay, ATM, MPLS etc. in order to transfer information over the WAN. The message sent from Portland campus would travel to the Hub, i.e. Kansas City, which will decide on which interface to forward the traffic so that it could reach the Miami campus. Each message contains all the information necessary to route it to the correct destination. The Hub router contains tables which have all the information of connected sites, their interfaces, IP addresses and so on. The Hub plays a vital role in this type of communication.

8. What would happen to the WAN’s connectivity if the Kansas City campus were to go offline? Explain your answer.

A hub site designed with a single WAN router introduces a single point-of-failure that will take down the entire WAN. As obvious from the above diagram, all the sites are connected with the Kansas City site. Kansas City Site router is responsible for end-to-end connectivity between all the sites. So, if it would go offline, WAN connectivity between all the sites would be lost.

9. What architectural changes would you propose to mitigate the risk covered in #8? Explain your proposal in detail.

As a star WAN configuration presents a risk with a single point-of-failure at the Hub site, it is recommended configuration to use a partial mesh configuration or even better is a full mesh configuration. If we plan to use a full mesh configuration, additional WAN links would have to be deployed which would add cost to the deployment. For this purpose, the following new WAN links need to be deployed:

  • From Portland Site to Los Angeles, Portland to San Antonio, Portland to Miami and Portland to New York City.
  • From Los Angeles to San Antonio, Los Angeles to Miami and Los Angeles to New York City
  • From San Antonio to Miami and San Antonio to New York City
  • From Miami to New York City

In this way we can eliminate the risk of single-point-of-failure by deploying a full-mesh WAN Network.

10. Now imagine that the organization is considering adding the following links to the original architecture illustrated in the diagram – Portland to New York City, New York City to Miami, Miami to San Antonio, San Antonio to Los Angeles, and Los Angeles to Portland.

a. What would be the advantages of adding these links? Explain your answer in detail.

Adding the above mentioned links to the existing WAN would make it a hybrid Configuration which contains both star and ring configuration. Hybrid configuration would provide the following advantages:

In hybrid networks, fault detection and troubleshooting is relatively easy. The part in which fault is detected can be isolated from the rest of network and required corrective measures can be taken, without affecting the functioning of rest of the network [5]. It’s easy to increase the size of network by adding new components, without disturbing existing architecture. The organization optimize the available resources and design the network according to the requirements in future. Hybrid configuration is the combination of two or more topologies, so it can be designed in such a way that strengths of constituent topologies are maximized while their weaknesses are neutralized. For example, Ring Topology has better reliability and Star topology has high fault-tolerance capability, so these two can be used effectively in hybrid star-ring topology.

b. What would be the disadvantages of adding these links? Explain your answer in detail.

One of the biggest drawback of hybrid topology is its design. It’s not easy to design this type of architecture and it’s a tough job for designers. Configuration and installation process needs to be very efficient. The hubs used to connect two distinct networks, are very expensive. These hubs are different from usual hubs as they need to be intelligent enough to work with different architectures and should be function even if a part of network is down. As hybrid architectures are usually larger in scale, they require a lot of cables, cooling systems, sophisticate network devices, etc. so the deployment cost of infrastructure increases dramatically.

 

References:

[1] Backbone network architectures (Data Communications and Networking) Part 1

http://what-when-how.com/data-communications-and-networking/backbone-network-architectures-data-communications-and-networking-part-1/

[2] Routed Vs. Switched Networks By James Woudon

http://www.ehow.com/info_12037417_routed-vs-switched-networks.html

[3] reasons for routed network vs a flat network https://supportforums.cisco.com/discussion/10146406/reasons-routed-network-vs-flat-network

[4] What Is a WAN Topology? by Michael Lawrence, Demand Media

http://smallbusiness.chron.com/wan-topology-68492.html

[5] Hybrid Topology: Advantages and Disadvantages of Hybrid Topology by Penna Sparrow

http://www.ianswer4u.com/2012/05/hybrid-topology-advantages-and.html#axzz3rpfVvixE

 

hihi


Want latest solution of this assignment

Want to order fresh copy of the Sample Template Answers? online or do you need the old solutions for Sample Template, contact our customer support or talk to us to get the answers of it.