A Small Network Design - Report Example

? A SMALL NETWORK DESIGN Table of Contents A small network design Introduction This report provides the requirementsand a possible network design for a building company which from its specifications has 14 branches with 13 users per branch. This means that there is a head office which co-ordinates the branches. The requirements include all the LAN devices and office automation devices. It also suggests the possible software which might be run in the servers, workstations and printers. The report also contains even finer implementation details e.g., the IP addressing scheme and suggestion for the actual sub netting. The costs of the local area network have also been included by providence of possible prices for the devices. Requirements The company requires a local area network to ensure efficient and effective communication, administration of the branches and it users. The main requirements here then are the devices to implement a local area network without concentrating on wide area network requirements. Some of the devices for a LAN include: routers for communication between branches, switches; for communication between devices in a LAN, cables; to interconnect devices, connectors; which ensure end to end connections between a cable and a device and of course office automation devices like servers e.g., file servers, mail servers and printers for storing and downloading files, for mail exchange and for installing printing software respectively. The company also will require software for their customized uses; the software might be outsourced or in-house. The company also requires operational staff software e.g., word processing. Other devices, which the company is ought to have includes telephony system such as using Voice over Internet Protocol (VoIP) and installing VoIP phones. VoIP is IP based and convergent, but use of other implementations like ISDN is possible. The company also requires mail servers e.g. Microsoft exchange or they might also build their own software which use protocols like IMAP and POP3. Another vital office requirement is printers and probably a printer server for scalability purposes. Devices and Cabling Cables and connectors Cables and connectors are necessary for devices to communicate with one another, some devices come with built in connectors such as network interface cards and wireless adapters. The company wants a general LAN design, and it might also decide to include wireless LAN, which uses connectionless media for devices to communicate but not advisable because it is prone to network intrusion such as phishing and spoofing. Almost all computers come with connectors like RJ45 for Unshielded Twisted Pair (UTP) connection. The main cable type, which will be used, is Unshielded Twisted Pair which are used for LANs. It has several categories according to performance. For this company, the ideal one is Category 5e Ethernet (cat5e). It has a jacket and an outside sheath. Cat5e is an extension of category 5 which has extra features to minimize cross-talk and signal interference, its speed is 125Mbps unlike Cat5 which is 100Mps. Depending on the number of users and need for fast connectivity the company might opt to use Category 7 of UTP. This is also called SSTP (shielded screen twisted pair). It has a data rate of 600Mps. Computers The company requires computers to carry out all the automated work in the company and also to install software for communication. It needs desktop computers for basic office automation software like word processing, emailing and accessing the internet. For software, which require more memory and processing power such as design software and working on graphics, the company will require work stations. Laptops also can be used for both small office work and also large software, a Core i5 laptop can be as powerful as a work station and some can be termed as mobile workstations. Switches Switches are intelligent devices, unlike repeaters and hubs and are suitable for computers in the LAN of the company to communicate with each other without any overload. They can handle two or more ports and can support communication between two or more devices. They run in duplex mode. When data transmission signal comes into one port, the switch finds the MAC address to which it can send the signal. Unlike a hub which broadcasts signals to all the devices, consuming bandwidth available, a switch identifies the MAC address of the device which the message is to be sent and only that device gets the request to accept the message. Switches will also allow room for VLAN (virtual local area network), which is a LAN configured by software, not by physical wiring. This requires installation of the switch with VLAN software. The advantage of VLAN is that it is possible to create a broadcast domain such as having a group of computers with one Multicast IP address where a signal can be multicast to that specific group. This will create virtual workgroups, reduce time and cost and security is improved. A switch also repeats signal on the necessary ports only contrary to a hub. Routers For the branches of the company to communicate with one another and also for the main-office to communicate with the 13 branches, a router is required. Of course, the company has to have a leased line with a local ISP (internet service provider). The ISP will provide the connections between the branches and this will be possible by having a router as Gateway in each branch. The router will allow for outside traffic to enter into the branch LAN, this traffic can be from the other branches. The router has a routing table, which it will keep all the IP addresses for the devices and forwards (routes) data to the correct branch. Servers Servers are powerful computers and are usually powerful CPUs stacked in a server cabinet. They run throughout and, therefore, are not interrupted unless during maintenance or if a restart is required. Due to their continuous running, they emit a lot of heat, and, therefore, require cooling. Servers will hold the server software which the users in the company will access, this include mail servers, printer servers and application servers. Multiple users can access the servers at the same time. They contain multiple CPUs and with powerful RAMs of 3 GHz or more. They are also tolerant to failure of software or hardware since they have redundant disks called Redundant Array of Inexpensive disks (RAID). The company can use one of the levels which are: RAID0, which has a minimum of 2 disks, does not have mirroring, Parity (error checking), and implements stripping of blocks of data hence efficient but should not be used for critical systems. RAID level I has no parity but has excellent redundancy due to mirroring of blocks; RAID level 2 uses bit level stripping across disks, uses hamming Error Correction Code, this level is no longer used as it is expensive; RAID level 3, uses byte level stripping and a separate dedicated disk for parity storing; other levels are 4, 5 which are not commonly used, RAID 6 uses block stripping and dual parity and expensive to implement in a RAID controller (Ramesh Natarajan, 2011). The best RAID level for mission critical systems is RAID level 10 also called RAID1+0 has a minimum of 4 disks, excellent redundancy due to mirroring of disks, excellent performance due to block stripping, but expensive. The servers should be kept in the head office in a special server room. The server room should only be accessed by the server administrators or authorised maintenance staff. The server room should also be secure and be backed up regularly or restarted when the services are down. Itemized costing Equipment description (cisco manufacturer) Cost Per Item (approx.) Modular Router w/2xFE, 2 WAN slots, 64 FL/256 DR $837 per router Cisco Catalyst 3750X-12S-E - switch - 12 ports - managed - rack-mountable $8,400 per switch Yellow Cable for Ethernet, Straight-through, RJ- 45, 6 feet $12 per cable Ethernet Cross-over Cable $15 each Cisco 12004 GSR AC Power Supply Cord, US $15 each HSSI Cable, Male-to-Male Connectors, 10 Feet $60 each High Density 4-port EIA-232 Cable, Male, DT $180 each Cisco 12404 220VAC Power Supply Cord, US $60 each TEO 2.1 Oracle Database Adapter Per Target Database $1908 each TES 3.0 Agt Linux Redhat Enterprise Server AS Rel 4 And 5 Per CPU $1113 each Cisco UC IP Phone 6961, Charcoal, Slimline Handset $177 each Arctic White Replacement Footstand for 6961 IP Phone $22 Computers – (2GB RAM, 3ghz with 320 GB hardisk) $400 per computer Monitors (15 inch) $60 per monitor Printer – (Color and LaserJet) $1000.00 per printer Total for one branch $14,254 Total for all branches $185,556 The costing of the items is dependent on the actual Kit Manufacturer decided upon by the company. The Kit Manufacturer who can be accessed easily or who have a good reputation can be approached. The itemized costing table above did not commit on one manufacturer in all cases but gave approximate price for each device. Physical topology IP addressing scheme IP is a communication protocol used both in the larger network and in the smallest organizational LANS. IP address is a unique identifier of a device given to a node (terminal device) in a computer network. It consists of four 8 bit numbers ranging from 0-255. Sub netting was initially developed because of the disadvantages posed by the two-level, addressing hierarchy (of network prefix and host number). Using one network number, the company can add subnets without requesting for another network number. Preferable addressing scheme The addressing scheme suitable for this company is Class C of the IPv4 addressing space. This is because; it does not waste much IP. A class B addressing scheme with network ID of 130.168.8.0 will have 2046 hosts per subnet but the company has only 13 users per subnet. This will waste the IP. The company thus is going to have subnets (sub-network) which are smaller networks within the larger LANs. Each of the 13 LAN branches in the company including the head office will be a single subnet with no further sub-divisions within it. This will make it easy for the switches in each branch to form a “broadcast domain” (McMillan, 2011). IP Address Network IP 192.168.3.0 Subnets Required 14 Subnet Host Required 13 Subnet Mask Classless inter-domain routing(CIDR) Notation /28 Subnet Mask Decimal 255.255.255.240 Subnet Mask Binary 11111111.11111111.11111111.11110000 Maximum Subnets 16 (2^4) Maximum Hosts Per Subnet 14 (subnets number -2) Subnet # Subnet IP Hosts Last IP(broadcast IP) 0 192.168.3.0 192.168.3.1 - 192.168.3.14 192.168.3.15 1 192.168.3.16 192.168.3.17 - 192.168.3.30 192.168.3.31 2 192.168.3.32 192.168.3.33 - 192.168.3.46 192.168.3.47 3 192.168.3.48 192.168.3.49 - 192.168.3.62 192.168.3.63 4 192.168.3.64 192.168.3.65 - 192.168.3.78 192.168.3.79 5 192.168.3.80 192.168.3.81 - 192.168.3.94 192.168.3.95 6 192.168.3.96 192.168.3.97 - 192.168.3.110 192.168.3.111 7 192.168.3.112 192.168.3.113 ­- 192.168.3.126 192.168.3.127 8 192.168.3.128 192.168.3.129 - 192.168.3.142 192.168.3.143 9 192.168.3.144 192.168.3.145 - 192.168.3.158 192.168.3.159 10 192.168.3.160 192.168.3.161 - 192.168.3.174 192.168.3.175 11 192.168.3.176 192.168.3.177 - 192.168.3.190 192.168.3.191 12 192.168.3.192 192.168.3.193 - 192.168.3.206 192.168.3.207 13 192.168.3.208 192.168.3.209 - 192.168.3.222 192.168.3.223 14 192.168.3.224 192.168.3.225 - 192.16.3.238 192.168.3.239 15 192.168.3.240 192.168.3.241 - 192.168.3.254 192.168.3.255 Alternative addressing scheme It is a fact that class network addressing scheme offers the capability and possibility of extra networks to be used. However, it does not provide for the possibility of adding more hosts within the sub networks. Therefore, class C addressing scheme is somewhat useless in the case of building a company’s network for long term purposes. Additional functionalities achieved through addition of network devices such as printers, servers, and VoIP telephones which would require new IPs would not be feasible and possible with class C sub-netting scheme. In this regard, it would be ideal to opt for an alternative addressing scheme such as class B, which offers enough network space and host that would ensure ease of scalability (McMillan, 2011). IP Address Scheme Network IP 172.16.8.0 Subnets Required 14 Subnet Host Required 13 Subnet Mask CIDR Notation /20 Subnet Mask Decimal 255.255.240.0 Subnet Mask Binary 11111111.11111111.11110000.00000000 Maximum Subnets 16 Maximum Hosts Per Subnet 4096 Subnet # Subnet IP Usable Host Range Broadcast IP 0 172.16.0.0 172.16.0.1 – 172.16.15.254 172.16.15.255 1 172.16.16.0 172.16.16.1 – 172.16.31.254 172.16.31.255 2 172.16.32.0 172.16.32.1 – 172.16.47.254 172.16.47.255 3 172.16.48.0 172.16.48.1 – 172.16.63.254 172.16.63.255 4 172.16.64.0 172.16.64.1 – 172.16.79.254 172.16.79.255 5 172.16.80.0 172.16.80.1 – 172.16.95.254 172.16.95.255 6 172.16.96.0 172.16.96.1 – 172.16.111.254 172.16.111.255 7 172.16.112.0 172.16.112.1 – 172.16.127.254 172.16.127.255 8 172.16.128.0 172.16.128.1 – 172.16.143.254 172.16.143.255 9 172.16.144.0 172.16.144.1 – 172.16.159.254 172.16.159.255 10 172.16.160.0 172.16.160.1 – 172.16.175.254 172.16.175.255 11 172.16.176.0 172.16.176.1 – 172.16.191.254 172.16.191.255 12 172.16.192.0 172.16.192.1 – 172.16.207.254 172.16.207.255 13 172.16.208.0 172.16.208.1 – 172.16.223.254 172.16.223.255 14 172.16.224.0 172.16.224.1 – 172.16.239.254 172.16.239.255 15 172.16.240.0 172.16.240.1 – 172.16.255.254 172.16.255.255 Logical topology The first IP address is always for the router and the last IP address in a subnet is broadcast IP. This is a logical diagram for one of the 14 branches (a subnet diagram). It also represents a broadcast domain since there are no further subnets within the small subnets created. Router (192.168.3.1) Switch 192.168.3.2 192.168.3.9 192.168.3.3 192.168.3.4 192.168.3.8 192.168.3.7 192.168.3.5 192.168.3.6 Broadcast Domain For hosts to communicate within the LAN, they use broadcast domains. Routers will prevent IP addresses, which are not for a specific subnet to access the network so only devices inside the subnet will communicate in the broadcast domain protected by router. For a host can only broadcast by using a broadcast packet. It addresses it to all the hosts in the network. The hosts communicate using MAC addresses, and there is no routing to other subnets or networks. For the company’s LAN design, all the subnets do not have further sub-nets and thus they become broadcast domains. Collision Domain By definition, a collision domain is a group of LAN devices whose frames can collide with one other. This happens with devices which are not intelligent since it is only one device which can be allowed to transmit at a time. Frames are lost if more than one devices try to transmit at the same time and the frames are irrecoverable and need to be resent. This will bring down the network, and it is also a security threat. In this particular case, all the group of devices in each of the branches form a collision domain since they communicate through a switch. The rate of collision domain is reduced in this case because we have used a switch a collision domain is not likely to be there unless hubs and repeaters are included which are not full duplex. Wireless network, which rely on an access point, can also result to a collision domain. Security There are many security issues which the company should be aware of, and can cause the company to lose millions of money if no proper mitigation measures are put in place. They include physical threats, electronic threats, technical failure of software of hardware and software and infrastructural failure. Physical threats Physical threats arise from damage of company servers and other IT resources either by fire, theft, or floods. It may also happen as a result of unauthorized access of information by malicious employees. The company servers should be put in a secure server room. Access should be restricted to authorized staff only. Electronic threats Such risks when they materialize aim at compromising the information security of a company. They usually come in the following ways: virus attacks such as love bug, which is a worm which spreads itself automatically through email or through chat rooms (Jesdanun, 2012) and Melissa viruses, can damage important files and replicate, filling up the available space in a server memory. For instance, it led Microsoft to switch off their mail server for some time (Rosencrance, 2002). Attach by black hat hackers who are malicious, and intrude on your IT system hoping to gain something out of the information. The company should use secret keys to encrypt information or it can opt for digital signature. The staff also misuses organizations emails and web. Organizations bandwidth can also be misused by Staff who performs peer-to-peer file transfer and sharing. Failure to update or configure the organization's software usually brings about these types of threats. Mitigating virus attacks can be done by purchasing an antivirus and installing in all the companies computers. They can then be scheduled to run at a specific time. Technical failure Occur when a hardware device ceases to function or some of its components gets or shows up due implementation errors. These failures can occur due to large, complex systems which are unstable and are frequently maintained and updated. Having maintenance staff can mitigate such failures and also outsourcing trusted software. Proper validation and verification of software can also be done before installing Infrastructural failure This occurs when there are network and power outages that render access to the network impossible. These will definitely interrupt the business process and valuable transactions might not take place. A company may have an Uninterrupted Power Supply for the purpose of mitigating power outages. However, in some occasions, the UPS may also fail (Kerans, 2008). Having redundant generators can help to prevent such scenarios in the company when it occurs because when one fails, another one is automatically switched on. Human error These happen when an employee fails to adhere or to be keen to IT stipulated policies and procedures or when an employee accidentally deletes important business data. It is, therefore, crucial for the company to provide policies and regulations on use of the company resources. Every employee when first employed should be given a copy of the IT policies of the company. Network Administration The network administrator installs, configures and maintains network devices. The network administrator also attends to any network problem which the users report. These include anything that might affect the whole company’s computer network or a mistake in the installation or support process. A network administrator thus should have a degree and above in computer science and several certifications under his belt e.g. CISA networking certificates. The network administrator also monitors or he/she himself installs important support programs to the company PCs. S/he should also manage accounts and give other staff the privileges to do some restricted access. He/she deletes or adds new employees to the network. A network administrator can be within the company staff, or he/she can also be outsourced. Another company’s services can be outsourced to administer the network if the staff are not competent enough. Reference Anon., 2006. Windows Networking. [Online] Available at: HYPERLINK "http://www.windowsnetworking.com/articles_tutorials/switch.html" http://www.windowsnetworking.com/articles_tutorials/switch.html [Accessed 2 May 2012]. Anon., 2009. Meyer Computers & Consulting. [Online] Available at: HYPERLINK "http://www.meyercomputer.net/Article-Hubs&Switches.html" http://www.meyercomputer.net/Article-Hubs&Switches.html [Accessed 2 May 2012]. Anon., 2011. Cat 5e network connectors. [Online] Available at: HYPERLINK " http://www.maplin.co.uk/cat-5e-network-rj45-connectors-32111" http://www.maplin.co.uk/cat-5e-network-rj45-connectors-32111 [Accessed 2 May 2012]. Anon., n.d. Subnetting Class C addresses. [Online] Available at: HYPERLINK "http://engweb.info/courses/itcn/labs/subnetting%20questions%20and%20tutorial.htm. " http://engweb.info/courses/itcn/labs/subnetting%20questions%20and%20tutorial.htm. [Accessed 2 May 2012]. Beasley, J.S., 2008. Networking. 2nd ed. Prentice Hall. Beveridge, D., 2000. Students Named in Love Bug Probe. [Online] Available at: HYPERLINK "http://www.cert.org" http://www.cert.org [Accessed 6 May 2012]. Ireland, J.T., 2005. Networking Made Easy. [Online] Available at: HYPERLINK "www.ictf.ox.ac.uk/conference/2005/presentations/wks11-networking.ppt." www.ictf.ox.ac.uk/conference/2005/presentations/wks11-networking.ppt. [Accessed 2 May 2012]. Kerans, P., 2008. How to Translate IT Risk Management into Competitive Advantage. [Online] Available at: HYPERLINK "http://www.eweek.com/c/a/Enterprise-Applications/How-to-Translate-IT-Risk-Management-into-Competitive-Advantage/2/" http://www.eweek.com/c/a/Enterprise-Applications/How-to-Translate-IT-Risk-Management-into-Competitive-Advantage/2/ [Accessed 2 May 2012]. Lammle, T., 2006. CCNA INTRO: Introduction to Cisco Networking Technologies Study Guide: Exam 640-821. New York: John Wiley & Sons. McMillan, T., 2011. Cisco Networking Essentials. New York: John Wiley & Sons. Natarajan, R., 2011. RAID 2, RAID 3, RAID 4, RAID 6 Explained with Diagram. [Online] Available at: HYPERLINK "http://www.thegeekstuff.com/2011/11/raid2-raid3-raid4-raid6/" http://www.thegeekstuff.com/2011/11/raid2-raid3-raid4-raid6/ [Accessed 6 May 2012]. Read More