CHAPTER 1: Network Fundamentals 26 on a bus network, it only listens or sends data. It doesn’t take data and then resend or regenerate it, so each computer isn’t essential to the network as a whole. If one computer fails, it doesn’t crash the entire network. Another benefit of this topology is that it is inexpensive, as less cable is used than other topologies we’ll discuss. As we’ll see later, some topologies have redundant connections or require a significant amount of cable. In a bus topology, every computer is connected to a single cable. Having a single cable, however, does cause other problems. If the cable breaks, then each segment has an end that isn’t terminated, and the entire network goes down. If the trunk is long enough, this can make it difficult to isolate where the break is. Another disadvantage of this topology is that it isn’t very scalable. The number of computers is limited to the length of the cable, and as your com- pany grows, it can be difficult changing the size and layout of the network. Also, if changes or repairs are made to the cable, the network is down because there is no redundancy and termination of the cable is required. STAR (Hierarchical) In a star topology, computers aren’t connected to one another, but are all connected to a central hub or switch. When a computer sends data to other computers on the network, it is sent along the cable to a central hub or switch, which can then pass the packets to the other computers or devices that are connected to it. As seen in Figure 1.4, when the computers are cabled to the hub, each point in the network can appear similar to points in a star (hence the name of this topology). Although this is an older topology that comes from the time when terminals were connected to mainframes as a centralized point, it is still a popular topology. Because cables run to a central point, if one cable breaks or fails in some way, only the com- puter that is connected to that cable is unable to use the network. Other computers are unaffected, because they have their own cables running to the hub. This can also make it easier to repair because, unlike the bus topology, it is easy to see where the cable failure has occurred. Another benefit of this topol- ogy is that it is scalable. As the net- work grows or changes, computers FIGURE 1.4 A Star Topology. Physical Networking Models 27 are simply added or removed from the hub. Unfortunately, because there is so much cabling that is being used to connect individual computers to a central hub, this also increases the cost of expanding and maintaining the network. Mesh A mesh topology has multiple connections, making it the most fault tolerant topology available. Every component of the network is connected directly to every other component. As seen in Figure 1.5, this creates a topology that provides redundant links across the network. If a break occurs in a segment of cable, traffic can still be rerouted using the other cables. In other words, if one connection fails, a computer can still access another computer or resource using another connection. Although it is almost impossible for a cable fault to crash a mesh topology, this topology is rarely used. There is significant cost and work involved in having network components directly connected to every other component. This topology provides redundant cable connections, but exponentially increases the workload and cost of maintaining them, making them difficult to manage and creating a cost that’s greater than other topologies. Rings As seen in Figure 1.6, a ring topology consists of computers connected to a cable that loops around. Because the topology is a closed loop, there are no uncon- nected ends to the ring, so terminators aren’t required. Data passes around FIGURE 1.5 A Mesh Topology. FIGURE 1.6 A Ring Topology Has All Computers Connected with a Cable That Loops Around. CHAPTER 1: Network Fundamentals 28 the loop in one direction. As it reaches each computer, the computer examines each packet and checks to see if any of the data packets are meant for it. If there aren’t any packets addressed to, the computer it sends the packets on to the next computer in the ring. In doing so, each computer acts as a repeater, resending the packet and thereby boosting the signal. When the packet reaches the originating computer, it removes the packet from the network. Ring topologies only allow one computer to have access to sending data on the ring, but provides equal access to the network. Bus and star topologies also allow only one workstation to communicate on the network at a time, but in a ring topology each computer is given a turn at having access. A signal called a token is passed from one computer to the next in the ring. When a computer has the token, it has access to the ring, and can send data. In a ring topology, if one computer fails, the entire network goes down. Because the topology requires an unbroken ring, if a computer is down or a cable is broken, the ring can’t be completed so the network can’t function properly. To avoid this, some rings have features that detect and disconnect failed computers from the ring, or will beacon and notify the network if a break is detected. By doing so, the network is able to function until the failed computer can be repaired. Point-to-Point A point-to-point topology is any network that connects two hosts in a dedi- cated fashion. For example, if you were to configure a router in New York to connect and use resources on a network in Atlanta, Georgia you would want to make sure you had a link between them that can support your needs. If you need a permanent connection that is constantly available and depend- able, you may need a T1 circuit. Although costly, you will be able to connect your two sites in New York and Georgia together without issue and have a point-to-point connection that is dependable and reliable. Point-to-Multipoint A point-to-multipoint topology is any network that connects three or more hosts and can grow exponentially based on the hardware and software you choose to manage it. For example, if you wanted to create a large network of many sites (i.e., New York, Georgia, Illinois, and Michigan), you may need to create a point to multipoint network. The main connection could be your headquarters location and the other three sites could be smaller sites accessing resources in the main hub site. This type of network is also called a hub and spoke topology. Physical Networking Models 29 Hybrid A hybrid topology is any mixture of at least two or more of any network topologies. So, it’s easy to say that any network that isn’t purely configured as one type of topology is configured as some form of hybrid network. Wireless A wireless topology broadcasts data over the air, so very few cables are used to connect systems together. As seen in Figure 1.7, this topology uses trans- mitters called cells, which broadcast the packets using radio frequencies. The cells extend a radio sphere around the transmitter in the shape of a bubble that can extend to multiple rooms and possibly different floors in a building. Each cell is connected to the network using cabling, so that it can receive and send data to servers, other cells, and networked peripherals. Computers and other devices have a device installed in them that transmits and receives data to and from the cell, allowing them to communicate with the network. Wireless networks can also extend their transmission to wireless devices by implementing radio antennas that are situated on buildings or towers. The antenna serves as a cell that will cover a wider area, such as a building or campus. This type of wireless network is particularly beneficial for pools of buildings that are close together and have some difficulty in connecting LANs in the buildings together using cables. Another method of wireless networking uses infrared communications, which requires a direct line of site and close proximity for the communica- tion to work. This type of wireless networking is similar to using a remote control for a TV, where each device needs to be lined up and within range of one another. Because of these limitations, it isn’t generally used for network- ing, but may be seen in a networked environment for connecting laptops and other computers to devices like printers. Because very little may be directly cabled together on a wireless network, there is a greater chance of transmis- sions being blocked or experiencing interference. Machinery and other devices can emit radio frequencies or electrical interference that disrupts signals being exchanged between FIGURE 1.7 A Wireless Network. CHAPTER 1: Network Fundamentals 30 the cell and wireless devices. Also, some buildings using cinderblocks, large amounts of metal or insulation to prevent transmissions from interfering with equipment can keep a wireless network from working between rooms. You may have experienced a blockage like this when you tried using a cell phone in certain buildings, and found it didn’t work. Signals are prevented from passing through these materials, and may require different frequencies to be used. EXERCISE 1.1 Understanding Types and Topologies The Freedom Fry Corporation has a network that’s contained within a single building. A foreign company that is located in Paris has bought them and would like to share data with them. The Paris firm has a similar network and it consists of several servers and computers networked together in a single building. However, each of the computers in the Paris firm is connected to a single hub, while the Freedom Fry Corporation’s client computers don’t use cables. Servers are connected to cables, which then branch off to hubs that connect to cells. On the basis of the information provided, answer the fol- lowing questions: What is the purpose of a WAN, and how could it be applied to this 1. network? What topology does the Paris firm use?2. What topology does the Freedom Fry Corporation use?3. On both networks, which networking component will be most 4. likely to fail? Exercise Answers A WAN allows an organization in diverse geographical locations to 1. be connected, and function as if they were part of the same LAN. It can be applied to this network by implementing a high-speed con- nection between the two offices. The Paris firm uses a star topology, which has computers of the 2. network all connected to a single hub. The Freedom Fry Corporation uses a wireless topology in which 3. the servers, hubs, and cells are all connected with cables, but computers use wireless adapters to connect to the cells. Network Types 31 Cables. Both networks use cables, which are a common point of 4. failure. Each of the other network components has less chance of failing than a damaged or faulty cable. NETWORK TYPES Just as we saw the Internet evolve from a relatively small network named ARPANet, networks can extend beyond their initial creation of a few comput- ers connected together. A network can be in a single building, or comprised of computers connected together over a broader geographical area. To categorize the scope of a network, different terms have been created to classify these dif- ferent network types. The types of networks that could be created include: Local area network Wide area network Metropolitan area network Storage area network Personal area network Campus area network LAN and WAN LANs and WANs were the first types of networks to be classified by the area they covered and are still the ones most commonly referred to. Although each of the names refer to an area, an exact range has never been firmly established and is left vague. Although IEEE (which we’ll discuss later in this chapter) defines a local area as being up to 4 km, no one will accuse it of not being a LAN if it is slightly over that. LANs are networks spanning a limited distance, while a WAN is a network that is larger than a LAN. What distinguishes a LAN from a WAN in terms of area is ambiguous and speculative at best. LANs are small- to medium-size networks, and generally connect net- work devices that are no more than a few miles of one another. They include networks that have been set up in homes, offices, the floor of a building, an entire building, a campus or group of nearby buildings, or facilities that are relatively close to one another. Basically, if you can walk or drive the distance of the network in a short time, you’re dealing with a LAN. CHAPTER 1: Network Fundamentals 32 Another way to characterize a LAN is through ownership. Typically, the network is owned by a single person or organization and is managed by a single person or group of people. For example, your home network would be a LAN that’s owned and managed by you. In the same way, a large company with several buildings in a region that’s run by a network administrator or IT department would also be a LAN. When you look at LANs in this way, you can see that most networks are actually LANs. WANs can span great geographical distances, and connect different LANs together using high-speed solutions or telephone lines. A WAN may connect LANs in different cities, regions, states/provinces, or even countries. This is something we saw when we discussed the first WAN, ARPANet, which connected the LANs of several institutions in different cities together. Over time, the number of computers and networks connecting to it grew until it spanned the world and became the Internet. By internetworking individual LANs together, the LANs become parts of a WAN. When looking at WANs, ownership isn’t a defining factor. WANs are often owned and managed by more than one organization. Each LAN that is part of the WAN may be managed by individuals or IT departments, and either maintains their connection to the rest of the LAN or hires outside parties to perform that function. For example, in the case of the Internet, you may maintain your home network, but you hire an ISP to maintain the connection to the World Wide Web. In the same way, a company with offices in different cities may hire the phone company to maintain a T1 line that connects the network together. An effective way of understanding how a LAN is related to a WAN is to look at how they are connected and how data is sent. This may differ from organization to organization, as there are several different ways of getting data from a LAN to a WAN, including: Modem, which is a device that allows you to connect to other com- puters and devices using telephone lines. Generally, when a modem is mentioned, it refers to a dial-up modem (as opposed to the digital modems for other methods mentioned later). This type of connection is slow, and allows connections at a maximum of 56 Kbps (meaning that 56,000 bits of data can be sent or received per second). Integrated Services Digital Network (ISDN), which also sends data over the telephone lines but at higher speeds, up to 128 kbps but averaging at 64 kbps using an ISDN modem or router. Digital Subscriber Line (DSL), which sends data across telephone lines at speeds ranging from 1.5 Mbps (1.5 million bits per second) using a router or digital modem and configured phone lines. Network Types 33 Cable, which transmits the data across cable lines using the same lines used for cable television at speeds of up to 1.5 Mbps. Satellite, which transmits data to a satellite at speeds of up to 400 Kbps. T1 and T3, which are dedicated connections that provide extremely high speeds. A T1 line provides speeds of 1.544 Mbps, while a T3 line provides speeds ranging from 3 Mbps to 44.736 Mbps To illustrate the relationships between LANs and WANs, let’s look at a situation that may be familiar to you: sending e-mail to another person. Using the e-mail program on your computer, you would click the send but- ton, and the data (i.e. your e-mail) would be sent to the device responsible for sending it to the Internet. If you had a dial-up account to the Internet, it might be a standard modem. If you had a LAN in your home and sent e-mail, it might be sent to a network adapter and sent over a network. As we mentioned earlier, when a network is created, two or more com- puters are connected together. In a LAN, these computers are in the same locale, such as being in the same room. Although wireless networking will be introduced in the next section, it’s important to know that wireless can operate globally via satellite, or on your local LAN. This differentiates between a Wireless LAN (WLAN) and satellite communications. Regardless, each computer has a network adapter installed in it, which transmits and receives data through a network cable or using wireless technology. When your data is sent to the network adapter, it is broken up into smaller chunks called packets that can be sent more efficiently over the network. As such, your e-mail would be broken into smaller packets, which would then be put together by the computer receiving it. If you used DSL, these packets would be transmitted over your home LAN to a router that is used to connect to the Internet, and also used to connect different computers on your LAN together. In cases where network cable is used, one end of a cable would be plugged into a network adapter, and the other would plug into the router. Data is sent over the cable with information on its destination, and the router determines if it’s for a com- puter on the LAN or needs to be sent to the ISP who provides DSL to you. As you’re sending e-mail to someone who isn’t on your home network, the router would use the DSL connection to send it from your LAN to the ISP’s LAN. In doing so, it has gone beyond the boundaries of your local network, and has been transmitted over a WAN. When the ISP receives your e-mail, it also looks at where the data is des- tined. Because the ISP also has a LAN, it looks at whether the e-mail is des- tined for someone else who uses their service, a computer on their network, CHAPTER 1: Network Fundamentals 34 or another network connected to the Internet. Because you’re sending the e-mail to someone who uses a different ISP, it sends the e-mail over the Internet, which is a giant WAN, to be received by the other ISP’s e-mail server. When the other ISP receives the data, it will store the e-mail you sent on its e-mail server, until your friend dials into the Internet using a modem. Your friend’s computer connects to the ISP’s server, and then requests any e-mail that the server might have. This data is again broken into packets, and sent over the telephone line so that your friend’s modem can receive the data, and the computer can reassemble these packets and display them in your friend’s e-mail program. As you can see by this example, there are many different kinds of LANs and WANs that data may pass through. LANs may be as small as a couple of computers networked together, and a WAN may be as large as the Internet or as small as two LANs (yours and your ISP’s) interconnected together using routers. In each case though, the LAN consists of computers that are part of the same network, and the WAN consists of geographically dispersed LANs that are internetworked. MAN Although most people refer to a network in terms of being either a LAN or a WAN, there are other terms to further categorize a network. One such category is a MAN, which is an acronym for metropolitan area network (MAN). A MAN will generally cover a metropolitan area like a city, but this isn’t always the case. For example, if you lived in a small town, and had your LAN connecting to another LAN in a neighboring town, you could also refer to this as a MAN. When LANs are connected together with high-speed solu- tions over a territory that is relatively close together (such as several build- ings in a city, region or county), it can be considered a MAN. A MAN is a group of LANs that are internetworked within a local geographic area, which IEEE (an organization we’ll discuss later in this chapter) defines as being 50 km or less in diameter. Exam Warning Being that the Network+ exam is an exam on networks, it should come as no surprise that questions dealing with LANs and WANs will appear. Make sure that you know the difference between a LAN and a WAN, and that a WAN is a group of internetworked LANs. Other types of networks discussed below (MAN, SAN, CAN, and PAN) aren’t covered extensively on the exam. Specific elements of LANs and WANs are discussed throughout this book, and you will need to know them to pass the exam. Network Types 35 SAN A SAN is a storage area network, and it is used to connect storage devices together using high-speed connections. It is a segment of a network, and allows storage devices to be accessed by computers within the larger LAN or WAN. These storage devices consist of hard disks or other methods of storing data, and allow users of the network to view and/or save data to a centralized location. PAN A PAN is a personal area network, which is a wireless network that allows devices to exchange data with computers. Personal digital assistants (PDAs), cell phones, and other devices that someone can carry on their person and support this technology have a wireless transmitter in them. When they are within a certain distance of a receiver that’s installed on a computer, data can be exchanged between the computer and the device. Using a PAN allows you to do such things as update a calendar in a PDA, address book in a cell phone, and other tasks that are supported by the device. CAN A CAN is a campus area network, and refers to a series of LANs that are internetworked between several nearby buildings. This is a common type of network that’s used in organizations with facilities that are close to one another, such as when there is a pool of office buildings or a campus. It is larger than a LAN but smaller than a MAN. Note With virtualization technologies becoming increasingly popular the need for centrally accessible storage arrays has become increasing important. Test Day Tip It is wise to quickly review information dealing with the Network+ exam shortly before taking the exam itself. A fast approach to reviewing is to look over the Exam Watch information, Summary of Exam Objectives, and Exam Objectives Fast Track sections of this book. To make it easy reviewing items you have a problem remembering and that may appear on the exam, highlight or bookmark these items in the book, so you can review them at crunch time. They will provide a quick approach to re-examining important information. . together, the LANs become parts of a WAN. When looking at WANs, ownership isn’t a defining factor. WANs are often owned and managed by more than one organization. Each LAN that is part of the WAN may. part of the WAN may be managed by individuals or IT departments, and either maintains their connection to the rest of the LAN or hires outside parties to perform that function. For example, in. mentioned later). This type of connection is slow, and allows connections at a maximum of 56 Kbps (meaning that 56 ,000 bits of data can be sent or received per second). Integrated Services Digital