Design a 200 host network, split into 4 subnets using VLSM, we have summarized the IP addressing scheme that we have utilized. A detailed topology of our design. Relate our design to the OSI model, fully explaining the purpose and function of the OSI model. We will need to fully explain the OSI role of the first 4 layers in networking, whilst indicating the key devices that are used at each layer. We will explain how a single IP packet will transfers over the network in relation to the OSI model.
Network design solution.
Holku Industries currently have four networks divided over four separate departments; the hosts available over the four networks consist of. Sales 75 hosts, manufacturing 65 hosts, warehouse 40 hosts, administration 20 hosts.
Due to the current situation there is more network traffic, overwhelming the band width of the network. Due to this we have been asked to subnet a single private class B address to minimise broadcast traffic. At the moment the hosts available to each network would be 65,534 hosts. Imagine a broadcast packet to that many hosts. Or, imagine a company with many locations. They could borrow bits from the host id octets to create a custom subnet mask thus creating multiple subnets from one larger one. It may provide for better traffic management or merely better design.
Once we subnet our class B we would have less wasted hosts. The Private class B addressing scheme was set aside by the Internet assigned number authority (IANA) RFC-1597. For organization`s not requiring internet access, if Holku industries require internet access for future development a network addressing translation would be required (NAT device) (RFC 1631)
We are using a Class B addressing scheme. A class C would be sufficient for the hosts required .Whilst class A addressing would be far too large. And would cause further broadcast traffic.
Before variable length sub-netting has taken place
Unused Host space
After variable length sub-netting has taken place the networks would be viewed as.
Point to point
Summarise IP addressing.
The Summarisation of the Network IP will be: 172.20.0.0/23, summarisation meaning everything before the 23rd bit will be the same. To list all the networks and corresponding bits, is not required as the bits only change after the 23rd bit. (See appendix for full workings in Binary and decimal.)
We have allowed for the maximum amount of cabling for each host. Approx 90m as a maximum we are fully aware that in practice nodes would not be set so far apart, however without a correct original topology we can only estimate.
Why we have chosen the particular hardware.
The reason for the particular hardware we have chosen
ü We have chosen the 2621-XM router for the adaptability it provides. The router provides a second modulation port. This will allow many upgrades for future growth.
ü The Cisco catalyst 2950-24 switch gives the ability to connect 24 nodes and has 2x gigabyte connections which we are using to interconnect our switches
ü Cat 6 UTP 90 meters max, 10m for patching 100m in total.
Why a router?
Before Holku industries had a very large network, due to this broadcast was sent throughout the network causing increasing network traffic because of this we are using router to define each network and contain the broadcast traffic to each subnet. A brief description of what a router will do for Holku industries.
Routers do not forward broadcast traffic and therefore define broadcast domains
Advantages of Routers:
ü Routers reduce network traffic as they do not retransmit network broadcasts.
ü Routers connect different network media, such as Ethernet and Token Ring.
ü Routers may choose the best path possible across a network using dynamic routing techniques.
ü Switches improving network performance, routers are used to segment large networks.
Disadvantages of Routers:
ü Routers are more costly than multiple hubs or bridges.
ü Routers are slower than repeaters because they must analyze all data from the Physical to the Network layer.
ü Dynamic router as in Rip/Rip 2 may cause additional network traffic.
As mentioned in the disadvantages section regarding routers (Dynamic router communication will cause additional network traffic). Due to this fact I have chosen not to use Dynamic routing protocol and simply use static routing, due to the size of the network this would be the best option to use. For further development when the network becomes larger dynamic routing (rip 2) would be beneficial.
We have added a further switch to the warehouse network, to aid in redundancy to prevent loss of production if a switch fails. We have only utilised one switch with administration network due to it being the smallest network. We have used two routers to help aid in future internet connectivity, allowing two lines with modulation, for the internet. As you can see on our topology, cables have been run for redundancy from switch to switch.
Pro and Con Conclusion.
We have found that if we use one router which would be cost effective (Pro), but would not offer further future proofing due to module socket being taken up(Con). We also found if we utilise two routers instead of one router, and one of those routers fail, we could re-route traffic to help essential production and loss of revenue(Pro). We have opted for the 24 port switch. Due to it being cost effective(Pro), and we have also concluded that if we opted for the 48 port switch or 128 port switch, and one of the large capacity switches fail, then this would prevent a potential problem with loss of production/revenue(Con). Further cabling would be required due to the position of the larger capacity switches. (Con) A further (Pro), would be if Holku needed internet access, then they could utilise both routers with modulation to connect to the internet, this would also help if one router failed, production on order from the internet would not be halted.
The purpose and function of the OSI model.
Open system interconnection (OSI)
The OSI model is simply a reference model in which manufacture may adhere to, this allowing multiple manufactures adhering to the OSI, thus allowing multiple Network adapters to inter-operate with one another. For instance, a Mac pc utilising TCP/IP conforming to the OSI model, and a Windows pc conforming allows the ability to communicate between one another.
Each layer is assigned a particular task that solely performs at that layer. Different protocols are used at these different layers to perform that assigned task. The OSI model defines how the data flows down the seven layers from sending to receiving.
Transport layer 4
The transport layer is the first purely logical layer in the model. Layer 4 maintains the control and integrity of a communication session. This layer is responsible for providing mechanisms for multiplexing. The transport layer is also responsible for transporting data in a sequential manner. The Transport layer also reassembles data from higher-layer application and establishes the logical connection between the sending and receiving host on the network. TCP and UDP are introduced and used at this layer.
The important functions performed at the Transport layer to enable network communications are listed below.
ü Guaranteed data delivery.
ü Named resolution.
ü Flow control.
ü Error detection.
ü Error recovery.
The Network Layer 3
The network layer of the OSI model is responsible for moving packets between devices; it is used to determine which path a packet would take to reach its destination over multiple paths and intermediate devices.
One of the main functions performed at the network layer would be routing. (Rip, Ripv2, EIGRP, and OSPF). Data packets used to transport user data through the network. Protocols used to support data traffic, example of a routing protocol would be IP, IPv6. The network layer, sets up logical paths or virtual circuits for transmitting data packets from a source network to a destination network. It performs the following functions.
ü Packet switching and routing
ü Error detection
ü Congestion control
ü At the network layer you will find Routers and layer 3 switches.
The Data-Link Layer 2
Data link enables the movement of data over a link from one device to another, this is done by defining the interface with the network medium. The responsibilities of the data link layer. Layer 2 manages the order of the bits, packets and frame encapsulation. This is the layer where data packets are prepared for transmission by the physical layer, at the data link layer you will find Mac addresses, VLANs as well as WAN protocols such as frame relay and ATM. Wireless networking is seen at this layer.
The data link layer is divided into the following sub-layers.
ü Logical link control 802.2, (LLC) layers responsible for identifying Network layer protocols and then encapsulates them. A LLC header is added.
ü Data link layer you will find switches and bridges at this level.
The physical Layer 1
The physical layer is responsible for the physical communication between end stations. At the physical layer is communicates directly with the media by turning all the information passed down the OSI model, at this level it then converts the data into bits. This transmits the raw bit over a physical medium. In the form of electro-mechanical terms of voltage and wavelength
The details on the actual physical connection include.
ü Network connection types and what media is used. (NIC interface card)
ü Data encoding, analogue and digital signalling used.
ü Bit synchronisation .
How data moves over the Lan
Warehouse PC 1 Pings destination PC administration 1 IP 172.20.1.67
The packet has moved down from Layer 3 to layer 1 and been sent onto the medium to the next hop warehouse switch. Now the Inward layers of the switch.
The packet has now reached layer 2 of the inward process, it has now been passed to the outward process starting at layer 2.
The switch has processed the packet and has now sent it on to its destination forwarding it to the router.
The packet has now been processed inward from layer one to three and has now moved to the outward process of the router.
The packet has now been forwarded after being processed by the router; it is now on the medium continuing on to its destination.
The switch has processed the packet inbound and has passed the packet to the outbound process.
The switch has now processed the packet and has re sent the packet onto the medium to carry on to its destination.
The ping has now found the destination pc after transverses over our local network, the IP packet will now return and re- send over the LAN four times to complete the ping process. For a smaller process I have chosen this scenario that only moves through one router. To show a packet that moves over our LAN, from one router to the next, HDLC protocol would have been used. I have chosen not to use that Instance to shorten the process.
We have used Gigabyte connection from switch to switch to aid for better flow of traffic as we are using two routers and the disadvantage of this would be the processing of data at each router, this would slow the network down. So by using our gigabyte connection this would help speed the flow of traffic. This is why we have utilized this particular hardware and not just picked hardware for ease of this assignment, but taken time in perusing what each hardware device is capable of. Glen Drews-Power
* (TCP/IP the Computer encyclopaedia 1998) The Computer Language Company pages 104/125
* Author: - Todd Laemmle Cisco Certified Network Associate study guide (Published by Wiley 2007) Chapter 1,2,3,4 pages 1 to 76.
* Author:- Wendell Odom Cisco, CCENT/CCNA ICND1 official Exam certificate guide, second edition (Published by Cisco press) chapter 2 pages 17/39 chapter 12 pages 331/350
* CCNA network fundamentals version 4.0 curriculums Chapters 4, 5, 6,7,10 accessed online at www.netacad.net on the 18 December 2009 @ 18:00 hrs.
* Maqsood Ghafoor lecture notes Semester one.