Networking Basics A network is a group of computers, printers, and other devices that are connected together with cables. Information travels over the cables, allowing network users to exchange documents & data with each other, print to the same printers, and generally share any hardware or software that is connected to the network. Each computer, printer, or other peripheral device that is connected to the network is called a node. Networks can have tens, thousands, or even millions of nodes. Like most things, networks are are assembled according to certain rules. Cabling, for example, has to be a certain length, each cabling strand can only support a certain amount of network traffic, etc. The rules that govern how a network is set up is called its topology. The most popular topology in use today is called Ethernet, which consists of computers and peripherals cabled together in specific ways. Ethernet is relatively inexpensive, easy to set up and use, and very, very fast. Ethernet networks are categorized by how fast they can move information. Speed is expressed in megabits per second (or Mbps), where one "bit" is equal to 1/8th of a character, letter, or number. There are currently two Ethernet speed categories. Standard Ethernet operates at a fast 10Mbps, which is quick enough for most networking tasks. Fast Ethernet, by contrast, races along at a blistering 100Mbps, making it ideal for desktop video, multimedia, and other speed-hungry applications. The new technology behind Fast Ethernet, which was introduced in the beginning of 1995, is not readily compatible with standard Ethernet. Making the two "talk" with each other requires special equipment (see switching hub below) and some knowledge of internetworking. If you're building your first network, decide whether to go with standard or Fast Ethernet before you begin shopping around for network hardware and software. Unless you plan on using video, multimedia, or heavy graphics software, plan on using standard Ethernet. For more information on standard and Fast Ethernet, see the Cabling & Hubs section. Cabling Basics The two most popular types of network cabling are twisted-pair (also known as 10BaseT) and thin coax (also known as 10Base2). 10BaseT cabling looks like ordinary telephone wire, except that it has 8 wires inside instead of 4. Thin coax looks like the copper coaxial cabling that's often used to connect a VCR to a TV set. Which type of cabling is best for you? Thin coax and 10BaseT can both be used exclusively or together, depending on the type of network that you're putting together. Small networks, for example, may want to use 10BaseT cabling by itself, because it's inexpensive, flexible, and ideal for going short distances. Larger networks (usually with 10 or more computers) may use a thin coax backbone with small clusters of 10BaseT cabling that branch off from it at regular intervals. Network Adapters A network computer is connected to the network cabling with a network interface card, (also called a "NIC", "nick", or network adapter). Some NICs are installed inside of a computer: the PC is opened up and a network card is plugged directly into one of the computer's internal expansion slots. 286, 386, and many 486 computers have 16-bit slots, so a 16-bit NIC is needed. Faster computers, like high-speed 486s and Pentiums, often have 32-bit, or PCI slots. These PCs require 32-bit NICs to achieve the fastest networking speeds possible for speed- critical applications like desktop video, multimedia, publishing, and databases. And if a computer is going to be used with a Fast Ethernet network, it will need a network adapter that supports 100Mbps data speeds as well. If a PC lacks expansion slots (which is true with portable PCs), special network adapters are used. A PCMCIA network adapter connects a PC to a network if the PC has a credit card-sized PCMCIA expansion slot, while a pocket adapter connects a PC to a network through its printer port. Hubs The last piece of the networking puzzle is called a hub. A hub is a box that is used to gather groups of PCs together at a central location with 10BaseT cabling. If you're networking a small group of computers together, you may be able to get by with a hub, some 10BaseT cables, and a handful of network adapters. Larger networks often use a thin coax "backbone" that connects a row of 10BaseT hubs together. Each hub, in turn, may connect a handful of computer together using 10BaseT cabling, which allows you to build networks of tens, hundreds, or thousands of nodes.Like network cards, hubs are available in both standard (10Mbps) and Fast Ethernet (100Mbps) versions. Client Server vs. Peer-to-Peer Every network requires special software to control the flow of information between users. A Network Operating System, or NOS, is installed onto each PC that requires network access. The NOS is like a traffic cop that monitors the exchange and flow of files, electronic mail, and other network information. Network Operating Systems are usually classified according to whether they are peer-to-peer or client-server NOSs. Peer-to-peer NOSs like Windows 95 and Windows for Workgroups are best for home & small office use-- they're great for sharing applications, data, printers, and other localized resources across a handful of PCs. Client-server NOSs like Windows NT and NetWare are ideal for large-scale organizations that require fast network access for video, publishing, multimedia, spreadsheet, database, and accounting operations. Peer-to-Peer Networks A peer-to-peer network allows two or more PCs to pool their resources together. Individual resources like disk drives, CD-ROM drives, and even printers are transformed into shared, collective resources that are accessible from every PC. Unlike client-server networks, where network information is stored on a centralized file server PC and made available to tens, hundreds, or thousands client PCs, the information stored across peer-to-peer networks is uniquely decentralized. Because peer-to-peer PCs have their own hard disk drives that are accessible by all computers, each PC acts as both a client (information requestor) and a server (information provider). In the diagram below, three peer-to-peer workstations are shown. Although not capable of handling the same amount of information flow that a client-server network might, all three computers can communicate directly with each other and share one another's resources. A peer-to-peer network can be built with either 10BaseT cabling and a hub or with a thin coax backbone. 10BaseT is best for small workgroups of 16 or fewer users that don't span long distances, or for workgroups that have one or more portable computers that may be disconnected from the network from time to time. After the networking hardware has been installed, a peer-to-peer network software package must be installed onto all of the PCs. Such a package allows information to be transferred back and forth between the PCs, hard disks, and other devices when users request it. Popular peer-to-peer NOS software includes Windows 95, Windows for Workgroups, Artisoft LANtastic, and NetWare Lite. Most NOSs allow each peer-to-peer user to determine which resources will be available for use by other users. Specific hard & floppy disk drives, directories or files, printers, and other resources can be attached or detached from the network via software. When one user's disk has been configured so that it is "sharable", it will usually appear as a new drive to the other users. In other words, if user A has an A and C drive on his computer, and user B configures his entire C drive as sharable, user A will suddenly have an A, C, and D drive (user A's D drive is actually user B's C drive). Directories work in a similar fashion. If user A has an A & C drive, and user B configures his "C:\WINDOWS" and "C:\DOS" directories as sharable, user A may suddenly have an A, C, D, and E drive (user A's D is user B's C:\WINDOWS, and E is user B's C:\DOS). Did you get all of that? Because drives can be easily shared between peer-to-peer PCs, applications only need to be installed on one computer--not two or three. If users have one copy of Microsoft Word, for example, it can be installed on user A's computer--and still used by user B. The advantages of peer-to-peer over client-server NOSs include: No need for a network administrator Network is fast/inexpensive to setup & maintain Each PC can make backup copies of its data to other PCs for security. By far the easiest type of network to build, peer-to-peer is perfect for both home and office use. Client-Server Networks In a client-server environment like Windows NT or Novell NetWare, files are stored on a centralized, high speed file server PC that is made available to client PCs. Network access speeds are usually faster than those found on peer-to-peer networks, which is reasonable given the vast numbers of clients that this architecture can support. Nearly all network services like printing and electronic mail are routed through the file server, which allows networking tasks to be tracked. Inefficient network segments can be reworked to make them faster, and users' activities can be closely monitored. Public data and applications are stored on the file server, where they are run from client PCs' locations, which makes upgrading software a simple task--network administrators can simply upgrade the applications stored on the file server, rather than having to physically upgrade each client PC. In the client-server diagram below, the client PCs are shown to be separate and subordinate to the file server. The clients' primary applications and files are stored in a common location. File servers are often set up so that each user on the network has access to his or her "own" directory, along with a range of "public" directories where applications are stored. If the two clients below want to communicate with each other, they must go through the file server to do it. A message from one client to another is first sent to the file server, where it is then routed to its destination. With tens or hundreds of client PCs, a file server is the only way to manage the often complex and simultaneous operations that large networks require. Network Printing In client-server networks, network printing is normally handled by a print server, a small box with at least two connectors: one for a printer, and another that attaches directly to the network cabling. Some print servers have more than two ports--they may, for example, support 2, 3, or 4 printers simultaneously. When a user sends a print job, it travels over the network cabling to the file server where it is stored. When the print server senses that the job is waiting, it moves it from the file server to its attached printer. When the job is finished, the print server returns a result message to the file server, indicating that the process is complete. In the diagram below, the laptop client PC sends a job to the file server. The file server, in turn, forwards the job to the print server, which sends it to the laser printer when it's available. Any client on the network can access the printer in this fashion, and it's quite fast. The print server can be placed anywhere on the network, and a network can have more than one print server--possibly one in an office's accounting department, another in marketing, and so on. Print Servers are available for both client-server and peer-to-peer networks. They're incredibly convenient because they let you put a printer anywhere along your network even if there isn't a computer nearby. However, users often opt not to use a print-server with their peer-to-peer network. Why? Because every computer's resources are available to everyone on the network, Sally can print a job on John's printer--just as if Sally had a printer attached to her computer. In this example, the printer is attached to the computer on the right. When the PC on the left sends a job, it "thinks" that it is printing to a printer of its own. In actuality, the job travels over the network cables to the PC on the right, which stores and prints the job in the background. The user at the PC with the printer is never interrupted while his computer processes and prints the job transparently. Remote Access & Modem Sharing When a client-server network needs a gateway to the world, the network administrator usually installs a remote- node server, which serves up two functions: remote access and modem sharing. Most remote-node servers attach directly to the network cabling; they provide a bridge between the network, a modem, and a telephone line. Remote access allows users to dial into their home networks from anywhere in the world. Once a connection has been established over ordinary phone lines by modem, users can access any programs or data on the network just as if they were seated at one of its local workstations. Some remote access servers only provide access to a file server's disk drives. Others can provide access to both the file server and direct access to any PC's hard disk on the network. This saves time because it allows a remote user to communicate directly with any network user without having to go through the file server. Modem sharing lets local network users dial out from their individual network computers to access the Internet, bulletin boards, America On-Line, and more. After firing up their favorite communications software, local users establish a link with the remote-node server over the network, which opens up an outgoing telephone line. Users' individual PCs don't need modems, which is a big money saver--only a single modem & phone line are required for tens or hundreds of users. In the case of peer-to-peer networks, by contrast, every PC requires its own modem for access to the outside world. All About Cabling All About Cabling The two most popular types of network cabling are twisted-pair (also known as 10BaseT) and thin coax (also known as 10Base2). 10BaseT cabling looks like ordinary telephone wire, except that it has 8 wires inside instead of 4. Thin coax looks like the copper coaxial cabling that's often used to connect a VCR to a TV set. 10BaseT Cabling When 10BaseT cabling is used, a strand of cabling is inserted between each computer and a hub. If you have 5 computers, you'll need 5 cables. Each cable cannot exceed 325 feet in length. Because the cables from all of the PCs converge at a common point, a 10BaseT network forms a star configuration, or geometric design, when viewed from above. In the figure below, three computers are connected together with 10BaseT cabling and a hub. A 10BaseT hub is basically a box with a row of 10BaseT jacks. Most hubs have 5, 8, 12, or 16 jacks, but some may have more. Most hubs also have an uplink port, which is a special 10BaseT or thin coax port that allows the hub to be connected to either (1) other hubs, or (2) a thin coax backbone (see below for information on backbones). By uplinking multiple hubs together, you can add additional computers to your network whenever you need to. 10BaseT cabling is available in different grades or categories. Some grades, or "cats", are required for Fast Ethernet networks, while others are perfectly acceptable for standard 10Mbps networks--and less expensive, too. About 85% of the networks in the U.S. use standard unshielded twisted-pair (UTP) Category 5 10BaseT cabling because it offers a performance advantage over lower grades. If you are using a 10Mbps network, category 3 is fine. If you plan on building a Fast Ethernet network at some time in the future, it's best to install Category 5 cabling. 10BaseT Category What It's Used For ------------------------------------------------------- 5 Fast Ethernet (and everything below) 4 Networks other than Ethernet 3 10Mbps 10BaseT 2 Alarms, telephone voice lines 1 Unknown (not rated for anything specific) If possible, decide whether you'll be using standard Ethernet or Fast Ethernet technology before you begin building your network. If you're not sure which technology you'll eventually use, choose to install Category 5 cabling. Remember, Fast Ethernet network adapters and hubs are not directly compatible with each other. It is possible to have both 10Mbps and 100Mbps segments on the same network, provided you have a switching hub between them that allows them to communicate. Want to know more about 10BaseT wiring configurations? Check out our wiring guide. Thin Coax Cabling The geometric design that is formed when thin coax cabling is used is called a linear or backbone configuration. The reason for this is that thin coax is always arranged in a straight line of PCs, hubs, or other devices. Thin coax networks always require termination, which is the act of "plugging up" both ends of the network. Instead of inserting an incoming thin coax cable directly into a PC, a T- connector is inserted instead, splitting the network adapter's input port into two separate ports. One port receives an incoming network cable; the other receives an outgoing network cable. If the PC is at the end of the network chain, a terminator plug is inserted into the empty hole of the T-connector. Thin coax is only used with 10Mbps Ethernet networks. Fast Ethernet networks, which are 10 times faster than standard Ethernet, use category 5 10BaseT cabling. The figure below shows three PCs connected together in a backbone configuration. Note that the backbone has termination at both ends, and each "T" connector plugs directly into a PC, where it allows for an incoming and outgoing connection. The maximum length for any thin coax segment is 607 feet. Mixing 10BaseT & Coax Finally, thin coax backbones and 10BaseT cabling & hubs can be connected together to allow for a wide variety of expansion options. In the more complex example below, a thin coax backbone connects two 10BaseT hubs together, along with a computer in-between. Each hub, in turn, branches off to still more computers with 10BaseT cabling. Note that the ends of the thin coax backbone are properly terminated. How to Pick Cabling There are two things to consider when deciding on the type of cable to use for your network. 1. How many PCs do you want to link together? 2. How long (in feet) is your network going to be? The answers to these two questions will determine the cabling that's best for you, and whether or not you'll need a hub. Use thin coax cabling if you... ------------------------------------------------ have fewer than 10 PCs, don't have any portable computers, and don't plan to expand Use 10BaseT cabling with a hub if you... ------------------------------------------------ have 16 or fewer PCs within a 325 foot radius of each other, have portable computers, and/or you plan to expand Use both thin coax and 10BaseT together if... ------------------------------------------------ you have more than 16 computers, or the radius of your workgroup is more than 300 feet Common Problems & Solutions Here are some ways to avoid the most common cabling pitfalls that network installers face. Avoid Interference Network cabling can be run under floors, around office dividers, or over dropped ceilings. When planning your wiring layout, try to keep cables away from power outlets, florescent lighting fixtures, uninterruptable power supplies, and other sources of strong electromagnetic interference. Coiling up cables can also cause interference. Thin Coax Cabling When using thin coax cabling, you must always use a T-connector at each PC and termination at both ends of the network, even if you're only connecting a couple of computers together. 10BaseT Cabling When using 10BaseT cabling, you must use a hub--even if you're only networking 2 PCs together. Many first time networkers forego a hub and simply plug a 10BaseT cable between two PCs' network cards. Such an installation is guaranteed to either (1) not work, or (2) be unreliable. The Big Picture The Big Picture Decide on a peer-to-peer or client-server NOS. Choose one or the other depending on the size & complexity of your network. Peer-to-peer allows individual PCs to share each other's hard disks, printers, and other resources. It's perfect for small networks with 16 or fewer users, or for workgroups with one or more portable PCs. Popular packages include Windows 95 and Windows for Workgroups. Client-server NOSs, by contrast, can handle heavier and more complex traffic loads than peer-to-peer, and are designed for large networks or speed-critical applications like video and multimedia. Popular packages include Windows NT and Novell NetWare. Plan your cabling layout. After you've decided on a network operating system software package, diagram how your network will go together. Plan on running cables under floors, over ceilings, or around dividers. If you're installing a small network over a short distance, use 10BaseT with one or more hubs. If you'll be running long distances (the radius of the network is more than 325 feet), plan on using thin coax cabling, possibly in combination with 10BaseT hubs. Make a hardware & software checklist. Get each of the items shown below. Be especially careful when choosing a network adapter for your PCs. Get ones that support your PC's PCMCIA, parallel port, or internal bus slots, as well as your network software. They should have both 10BaseT and thin coax ports on-board for flexibility. And if you'll be using Fast Ethernet, be sure they support 100Mbps speeds. If you plan to install a 10BaseT hub, make sure that it's expandable, and that it has enough ports to service all of your PCs. If you'll be using the hub in conjunction with a thin coax backbone, make sure that it has both 10BaseT and thin coax ports on-board. Get the correct hub for the type of Ethernet network that you're installing--either standard 10Mbps Ethernet or Fast 100Mbps Ethernet. If you'll be joining standard and Fast Ethernet segments together, you'll need a switching hub in-between. After obtaining the network software, adapters, cabling, and hub(s) (if any), install the network cabling first, followed by the network adapters, and finally, the network software. Here's what you need: a peer-to-peer or client-server NOS software package a PCMCIA, pocket, or internal network adapter for each PC If you're using 10BaseT... one Category 3 or 5 10BaseT cable for each PC (max length: 325 feet) one expandable 10BaseT hub with enough ports to service all of your PCs If you're using thin coax... one RG58/U thin coax cable for each PC (max length: 607 feet) one T-connector for each PC two 50-ohm terminators (one for each end of the network) Introducing the Instant Ethernet Series from Linksys. Linksys is one of the world's leading manufacturers of networking hardware, with a complete line of high performance network interface adapters, hubs, print servers, and remote-access servers. Price, Performance, and Reliability Major industry publications like PC Week, PC Magazine, and Windows Sources agree that our hardware offers the best values in price/performance that networking dollars can buy. Every network product includes both 10BaseT & 10Base2 ports, high speed logic, buffering, extensive software suites, free technical support and software upgrades, and complete documentation. Easy Installations Our products are designed to get you running right out of the box. All of our network adapters, for example, require no hardware settings and are fully plug-and-play compatible and software-configurable. Just plug them in. Wide PC & Networking Operating System Support Linksys products are guaranteed to run with your network software and all of your computers. Every product comes ready-to-run with an amazing number of PCs and networks, including Windows 95, Windows for Workgroups, NetWare, Windows NT, LANtastic, Banyan Vines, LAN Manager, PC-NFS, TCP/IP, and most others. Old Fashioned Customer Service Linksys makes customer service its highest priority. Every product that we sell is backed by a full, limited warranty, free, unlimited technical support, and free software upgrades. Intro to Fast Ethernet Welcome to the Linksys Fast Ethernet information center! Whether you're building a network from scratch and want mind-bending speed, or if you're simply expanding an existing Fast Ethernet network, Linksys has a full line of screaming 100Mbps hardware that's ripe for video conferencing, multimedia, graphics, and other speed-intensive applications. Building a Fast Ethernet Network from Scratch Build a Fast Ethernet Network From Scratch Building a fast ethernet 100Mbps network is a cinch -- all you need are a few desktop or lap top network adapters, a fast ethernet hub, and a Category 5 network cable for each computer. To make things easier, your local retailer carries network starter kits that come with everything you need to connect two computers together. The Fast Ethernet Starter Kit from Linksys shown here, for example, includes two 32-bit desktop network cards, two Category 5 network cables, and a high speed 4-port hub, making it ideal for home or small office use. EXPANSION MADE EASY Expanding a Fast Ethernet network amounts to nothing more than plugging in additional hubs and network cards. Fast Ethernet networking rules allow no more than two hubs to be connected together in one area. This is called the two-hub rule. This is fine if you don't plan on expanding too much; you could connect two 24-port hubs together, for example, and you'd be set to handle up to 48 users. A standard Fast Ethernet hub is perfect for this task However, as your needs grow, you might find yourself needing to handle fifty, sixty, or even hundreds of users! Fast Ethernet's solution for this problem is the stacking hub. A stacking hub is a specially designed hub that can "stack" to other hubs just like it. Here's the trick: once a stack of stackable hubs have been connected together, they act as a single hub, effectively fooling the network into thinking that only one hub is present! The expansion possibilities are enormous! For example, if you connect six stacking hubs This entire stack together, and each hub has ten ports, you're given a total of sixty ports--and of stackable hubs the stack acts as one hub, which doesn't violate Fast Ethernet's 2-hub rule. is acting as a single hub! When considering a new hub for your Fast Ethernet network, consider how much you're going to need to grow in the future. If you plan on extensive growth, choose a stackable hub so you can add other hubs to it without reaching the 2-hub rule. If, on the other hand, you only expect modest network growth in the near future, a standard Fast Ethernet hub will more than meet your needs. How to Speed Up an Existing Fast Ethernet Network How to Speed Up an Existing Fast Ethernet Network Fast Ethernet networks normally operate with shared bandwidth, which means that the network's overall speed is spread out across the available number of nodes. If you have a hub with 5 nodes attached to it, and the total bandwidth of the network is 100Mbps, then each node receives 100/5, or 20Mbps, of bandwidth. If you have only a few nodes, shared bandwidth isn't a problem. However, imagine spreading 100Mbps across 30 nodes -- even at Fast Ethernet speeds, the network could slow to a crawl during peak traffic periods. This 5-Port switch Fast Ethernet's answer to the problem of shared bandwidth is a Fast Ethernet switch. A is small enough switch transforms a shared network into a switched network, whereby each node gets to fit on any access to the network at the maximum bandwidth speed. If you have five ports, for example, small office desk. each port receives 100Mbps access, provided you have a switch installed. Switches also speed up network access because of their advanced addressing technology. In a shared network, data packets bounce around the network until they finally find their destination, which is not particularly efficient. Switches, on the other hand, examine each data packet individually, and direct them only to their intended destinations. The difference between shared and switched networks is like the difference between a shotgun and a laser. Not only can switching improve a Fast Ethernet network's efficiency, but when used in conjunction with a 10BaseT network, it can reduce wasted bandwidth there as well, sometimes with performance increases as high as 40%. Switches come in all shapes and sizes. Just like hubs, they are available in both desktop and rack mountable versions that can be integrated with both network adapters and hubs effortlessly. If your network demands optimum speed for video, multimedia, graphics, or other intensive applications, the extra boost that switches provide might be well worth the purchase. What is a Switch Without a switch installed, a network can get bogged down quickly as traffic rises. Traffic jams happen because data is forced to wander the entire network in search of its destination. A switch corrects traffic jam problems by ensuring that data goes straight from its origin to its proper destination, with no wandering in-between. Switches remember the address of every node on the network, and anticipate where data needs to go. Nodes connected to a switch can expect an immediate 40%-60% increase in performance. A switch can also connect networks of different speeds together. A 100Mbps network, for example, could be connected to a slower 10Mbps network by inserting a switch between the two networks. In this way, switches are good for migrating to faster network speeds without having to discard older legacy network hardware. This EtherFast Dual-Speed 10/100 16- IS MY NETWORK A GOOD Port Switch is the ideal centerpiece for CANDIDATE FOR A SWITCH? any high-performance network. If you do more than simple file and printer sharing, you should definitely consider a switch. Switch prices have fallen drastically since fall of 1998, and many are priced only slightly higher than regular hubs -- and since most hubs can't offer the performance benefits of switching, buying a switch is a smart move for any network, even if you have only a few users. In short, if your network needs maximum bandwidth and speedy performance, buy switches instead of hubs. In the picture above, the switch ties a file server, a high-powered PC, a print server, and a hub together for maximum bandwidth. The switch gives the hub (and hence, the workgroup connected to it) extremely fast access to the print server and file server. Access could be improved even more if the hub was replaced with another switch. In short, if your network needs maximum bandwidth and speedy performance, buy switches instead of hubs. WHAT KIND OF SWITCHES DOES LINKSYS OFFER? Linksys offers a full line of 10/100 switches. These high-powered performers are built to run with 10BaseT or 100BaseTX networks, or both. Drop one in to speed things up, or use a Linksys switch to connect a 10Mbps network to a faster 100Mbps LAN. Linksy switches are available in 2, 4, 6, 8, 10, 12, 16, and 24-port models for desktop and/or rack mounting use. Choose a desktop switch for home, small office, or departmental use where speed increases are needed. Rack mountable switches are best for wiring closets and enterprise deploymen where Internet, LAN, and internetworking connections need all of the speed they can get! A Strictly Fiber Diet 100BaseFX fiber cabling is the ultimate in speed-intensive incredible levels of bandwidth, expandability, and data integrity over distances of up to 6500 feet! Get your fiber here! Click on any of the links at right to learn more. Intro to Fiber Networking Introduction to Fiber Networking Fiber cabling is the best way to get the ultimate in speed-intensive performance from your 100Mbps Fast Ethernet network. Also known as 100BaseFX, fiber optics can serve up incredible levels of bandwidth, expandability, data integrity that can span distances of up to 2000 meters (6560 ft.) without signal degradation. Linksys offers 100BaseFX add-on modules for many of its EtherFast 10/100 Switches, StackPro II 10/100 Dual-Speed Hubs, 10/100 Auto-Sensing Hubs, and Fast Ethernet hubs. The modules can be used to uplink your hardware to high-speed fiber backbones, or allow communication between hubs and switches across great distances. Keep reading to find out which add-on module is best for your fiber networking needs.<> Fiber is commonly used for: Long hauls of a thousand feet or more between buildings or LANs, especially in school, university, and business campus settings. Fiber serves as a backbone between hubs that, in turn, branch off to workgroups with hubs and twisted-pair cabling High-bandwidth applications like video or multimedia Gigabit (1000Mbps) networking Fiber optic networking requires a great deal of planning before the cabling is installed. Factors such as noise loss, connection type, and cabling distance must be calculated in order to ensure a fail-safe networking environment. If you have never worked with fiber before, seek the help of a professional who is familiar with the medium. What About the Cabling? Like thin coax, fiber optic cabling is used primarily for network backbones. Made from flexible, optically efficient strands of glass coated with a layer of rubber tubing, fiber uses photons of light instead of electrons for sending and receiving data. Although fiber is physically capable of carrying Terabits (!) of data per second, the signaling hardware on the market today can handle no more than a few gigabits of information per second. Fiber cables come in different shapes and sizes with different connector types. The most commonly used fiber optic cable is a multi-mode fiber cable (MMF) with a 62.5 micron fiber optic core -- this is the kind of cabling that all Linksys products use. Single-mode fiber, another kind of cabling, is more efficient than multi-mode, but far more expensive. Fiber network segments always require two fiber cables: one for transmitting data, and one for receiving. Each end of a fiber cable is fitted with a plug that can be inserted into a network adapter, hub, or switch. In the United States, most cables use a square SC connector that slides and locks into place when inserted into a node or connected to another fiber cable; Europeans use a round ST connector instead. Since light tends to dissipate quickly when moving from one material to another, keeping the number of connections, or hops, along a network path to a minimum is essential. The more connectors you have and the longer your fiber link cable is, the higher the optical loss will be. Optical loss is measured with fiber optic test instruments that can tell exactly how much optical loss there may be on a given segment at a given wavelength of light. A standard grade fiber optic cable operating at 850 nms (nanometers per second) will have something in the neighborhood of from 4 dB to 5 dB loss per 1000 meters. SC or ST Connectors? First, you need to decide which type of fiber connectors you will be using. SC-type connectors are square in shape and are most common in North America. ST-type connectors are round and are more common in Europe and some parts of Asia. Both connectors offer the same features as far as distance and reliability, but it is best to choose one type of connector and stick with it over your entire network. Connectors of different types can communicate with the use of adapters or couplers. Whenever you need to change a connector type or if you require any SC-ST conversion adapters, always seek the help of a professional, as special tools need to be used. Full Duplex or Half Duplex? Unlike an RJ-45 port, a single fiber port is actually divided into two separate uni-directional ports. One port sends data and the other port receives data. Some fiber add-on modules are capable of sending and receiving data at the same time. This is known as full duplex (FDX) operation. Other add-on modules are only capable of either sending or receiving data at any given time and cannot do both at once. This is known as half duplex (HDX). As you might guess, FDX is the faster of the two. If you are adding fiber capabilities to an EtherFast 10/100 Switch, a StackPro II 10/100 Dual-Speed hub, or other device capable of handling a full duplex connection, you should always go with a full duplex fiber add-on module. If you are using a fiber add-on module on a 10/100 Auto-Sensing Hub, Fast Ethernet hub, or other fixed half duplex device, then you fiber optic options will be limited to half-duplex connections. Of course, a half duplex connection will not allow the distance of a full duplex connection. If long distance is of great importance to you, you should consider upgrading your hardware to allow support for a full duplex connection. Also, it is always b est to match the duplex operation of any two devices that are communicating with one another. When a full duplex device tries to communicate with a half duplex device, unstable network connections will result. Distance Extender or Transceiver? Fiber optic distance extender add-on modules can send data much farther than transceiver add-on modules. Some hub models are only capable of communicating with a transceiver, and transceivers only run at half duplex. If your device is compatible with a distance extender, always choose a distance extender first. How Far Can My Connection Go? This chart shows the precise measurements for fiber optic cabling distances. Type of Distance Distance Distance Extender to Distance Extender (FDX) 2000 meters (6560 feet) Distance Extender to Distance Extender (HDX) 412 meters (1330 feet) Distance Extender to Transceiver (HDX) 208 meters (672 feet) Transceiver to Transceiver (HDX) 16 meters (50 feet) Notes: Module type refers to whether a module is a transceiver or a distance extender. Most modules are distance extenders, which can send & receive data over distances of thousands of feet. Transceivers have shorter ranges. To learn more, see our Introduction. All ports labeled fiber are 100BaseFX. All modules support both half and full duplex unless otherwise noted; modules with support for both can be configured via on-board jumpers. Remember that each fiber run between two points uses two fiber cables: one for transmitting and one for receiving. If a fiber module above is listed as having 1 fiber port, it actually means that the module has two fiber connectors -- one for transmission, and one for reception -- in order to make a single fiber run.