What is Network?
“A network is defined as two or more computers linked together for the purpose of communicating and sharing information and other resources”.
Basic Requirements of a Network:
Connections include the hardware (physical components) required to hook up a computer to the network. Two terms are important to network connections:
The network medium: The network hardware that physically connects one computer to another. This is the cable between the computers.
The network interface: The hardware that attaches a computer to the network medium and acts as an interpreter between the computer and the network. Attaching a computer to a network requires an add-in board known as a network interface card (NIC).
Communications establish the rules concerning how computers talk and understand each other. Because computers often run different software, in order to communicate with each other they must speak a "shared language." Without shared communications, computers cannot exchange information, and remain isolated.
A service defines those things a computer shares with the rest of the network. For example, a computer can share a printer or specific directories or files. Unless computers on the network are capable of sharing resources, they remain isolated, even though physically connected.
Types of Network:
There are two types of network.
LAN (LOCAL AREA NETWORK):
A LAN (local area network) is a network that covers a limited distance (usually a single site or facility) and allows sharing of information and resources. A LAN can be as simple as two connected computers, or as complicated as a large site. This type of network is very popular because it allows individual computers to provide processing power and utilize their own memory, while programs and data can be stored on any computer in the network. Some of the older LANs also include configurations that rely totally on the power of a mini or mainframe computer (a server) to do all the work. In this case, the workstations are no more than "dumb" terminals (a keyboard and a monitor). With the increased power of today's personal computer, these types of networks are rare.
WAN (Wide Area Networks):
A wide area network (WAN) spans relatively large geographical areas. Connections for these sites require the use of ordinary telephone lines, T1 lines, ISDN (Integrated Services Digital Network) lines, radio waves, or satellite links. WANs can be accessed through dial-up connections, using a modem, or leased line direct connection. The leased-line method is more expensive but can be cost-effective for transmission of large volumes of data.
Physical layout of network is called network topology.
There are three types of network.
Ring Topology Bus Topology Star Topology
In a star network, all devices are connected to a central point called a hub. These hubs collect and distribute the flow of data within the network. Signals from the sending computer go to the hub and are then transmitted to all computers on the network. Large networks can feature several hubs. A star network is easy to troubleshoot because all information goes through the hub, making it easier to isolate problems.
In a bus network, all devices are connected to a single linear cable called a trunk (also known as a backbone or segment). Both ends of the cable must be terminated (like a SCSI bus) to stop the signal from bouncing. Because a bus network does not have a central point, it is more difficult to troubleshoot than a star network. A break or problem at any point along the bus can cause the entire network to go down.
In a ring network, all workstations and servers are connected in a closed loop. There are no terminating ends; therefore, if one computer fails, the entire network will go down. Each computer in the network acts like a repeater and boosts the signal before sending it to the next station. This type of network transmits data by passing a "token" around the network. If the token is free of data, a computer waiting to send data grabs it, attaches the data and the electronic address to the token, and sends it on its way. When the token reaches its destination computer, the data is removed and the token sent on.
Network cable and connectors:
All networks need cables. The three main types are twisted-pair cable (TP), coaxial cable, and fiber-optic cable (FDDI—Fiber Distributed Data Interface).
Twisted-pair cable, shown in the picture consists of two insulated strands of copper wire twisted around each other to form a pair. One or more twisted pairs are used in a twisted-pair cable. The purpose of twisting the wires is to eliminate electrical interference from other wires and outside sources such as motors. By twisting the wires, any electrical noise from the adjacent pair will be canceled. The more twists per linear foot, the greater the effect.
Twisted-pair wiring comes in two types: shielded (STP) and unshielded (UTP). STP has a foil or wire braid wrapped around the individual wires of the pairs; UTP does not. The STP cable uses a woven-copper braided jacket, which is a higher-quality, more protective jacket than UTP.
Twisted pair cable
Of the two types, UTP is the most common. UTP cables can be further divided into five categories:
Category 1: Traditional telephone cable. Carries voice but not data.
Category 2: Certified UTP for data transmission of up to 4 Mbps (megabits per second). It has four twisted pairs.
Category 3: Certified UTP for data transmission of up to 10 Mbps. It has four twisted pairs.
Category 4: Certified UTP for data transmissions up to 16 Mbps. It has four twisted pairs.
Category 5: Certified for data transmissions up to 100 Mbps. It has four twisted pairs of copper wire.
Twisted-pair cable has several advantages over other types of cable (coaxial and fiber-optic)—it is readily available, easy to install, and inexpensive. Among its disadvantages are its sensitivity to EMI (electromagnetic interference) and susceptibility to eavesdropping; it does not support communication at distances of greater than 100 feet; and it requires the addition of a hub (a multiple network connection point) if it is to be used with more than two computers.
CAT5 Cabling Issues
Ethernet networks use unshielded twisted pair (UTP) Category 5 cable. CAT5 cable runs should not exceed 100 meters.
Coaxial cable is made of two conductors that share the same axis; the center is a copper wire that is insulated by a plastic coating and then wrapped with an outer conductor (usually a wire braid). This outer conductor around the insulation serves as electrical shielding for the signal being carried by the inner conductor. Outside the outer conductor is a tough insulating plastic tube that provides physical and electrical protection. At one time, coaxial cable was the most widely used network cabling. However, with improvements and the lower cost of twisted-pair cables, it has lost its popularity.
Coaxial cable is found in two types: thin (ThinNet) and thick (ThickNet). Of the two, ThinNet is the easiest to use. It is about one-quarter of an inch in diameter, making it flexible and easy to work with (it is similar to the material commonly used for cable TV). ThinNet can carry a signal about 605 feet (185 meters) before the signal strength begins to suffer. ThickNet, on the other hand, is about three-eighths of an inch in diameter. This makes it a better conductor—it can carry a signal about 1,640 feet (500 meters) before signal strength begins to suffer. The disadvantage of ThickNet over ThinNet is that it is more difficult to work with. The ThickNet version is also known as standard Ethernet cable.
When compared to twisted-pair, coaxial cable is the better choice even though it costs more. It is a standard technology that resists rough treatment and EMI. Although more resistant, it is still susceptible to EMI and eavesdropping.
Use coaxial cable if you need:
A medium that can transmit voice, video, and data.
To transmit data longer distances than less-expensive cabling.
A familiar technology that offers reasonable data security.
A Mixed-Cable System
Many networks use both twisted-pair and coaxial cable. Twisted-pair cable is used on a per-floor basis to run wires to individual workstations. Coaxial cable is used to wire multiple floors together. Coaxial cable should also be considered for a small network because you can purchase prefabricated cables (with end connectors installed) in various lengths.
Fiber-optic cable is made of light-conducting glass or plastic fibers. It can carry data signals in the form of modulated pulses of light. The plastic-core cables are easier to install, but do not carry signals as far as glass-core cables. Multiple fiber cores can be bundled in the center of the protective tubing.
When both material and installation costs are taken into account, fiber-optic cable can prove to be no more expensive than twisted-pair or coaxial cable. Fiber has some advantages over copper wire; it is immune to EMI and detection outside the cable and provides a reliable and secure transmission media. It also supports very high bandwidths (the amount of information the cable can carry), so it can handle thousands of times more data than twisted-pair or coaxial cable.
Cable lengths can run from .25 to 2.0 kilometers depending on the fiber-optic cable and network. If you need to network multiple buildings, this should be the cable of choice. Fiber-optic cable systems require the use of fiber-compatible NICs.
There are two different types of RJ-45 connectors. There is the "bent tyne" connector intended for use with solid core CAT5, and then there is the "aligned tyne" connector for use with stranded CAT5 cable. Errors have popped up when using incorrect cable/connector combinations. The "bent tyne" connector will work just fine on stranded wire by the way, just not the other way around. In general, make sure your connector matches you cable type...
Standards set forth by EIA/TIA 568A/568B and AT&T 258A define the acceptable wiring and color-coding schemes for CAT5 cables.
ST or Straight Connector
MTRJ Connector MTP Connector
Duplex SC Connector
There are two common types of fiber optic connectors: SC and ST. The ST or "straight tip" connector is the most common connector used with fiber optic cable, although this is no longer the case for use with Ethernet. It is barrel shaped, similar to a BNC connector, and was developed by AT&T. A newer connector, the SC, is becoming more and more popular. It has a squared face and is thought to be easier to connect in a confined space. The SC is the connector type found on most Ethernet switch fiber modules and is the connector of choice for 100Mbit and Gigabit Ethernet. A duplex version of the SC connector is also available, which is keyed to prevent the TX and RX fibers being incorrectly connected.
There are two more fiber connectors that we may see more of in the future. These are the MTRJ and MTP. They are both duplex connectors and are approximately the size of an RJ45 connector.
Network Interface Cards:
Network interface cards (NICs) link a computer to the network cable system. They provide the physical connection between the computer's expansion bus and the network cabling.
Installation of the network interface card is the same as for any other expansion card. It requires setup of the system resources: IRQ, address, and software. Most cards today allow connection
for either thin Ethernet or UTP (unshielded twisted-pair) cabling. Thin Ethernet uses a round BNC connector, and UTP uses a RJ-45 connector (similar to a telephone jack).
Network interface card
Installing a NIC is just like installing any other expansion card. If you are installing a Windows 95-compliant Plug and Play card in a Windows 95 or Windows 98 machine, you'll simply need to physically install the card and boot up the computer. The card will be detected and, more than likely, install itself. You might only need to answer a few questions along the way. It requires a little more work to install a NIC in an operating system that is not Plug and Play-compliant. Installing network cards includes the following steps:
Be sure to document any changes that you make to the existing computer. This will eliminate any confusion in the installation process and provide future reference in case of problems.
Determine whether the card needs IRQ, DMA (direct memory access), or address settings. Remember that you might have to configure these manually, so be sure to check the card's documentation for default settings and instructions for how to make any needed changes.
Determine whether the necessary settings are available on the machine on which they will be installed. If proper documentation is not available, use diagnostic software such as Microsoft Diagnostics (MSD) to determine settings. Also check your AUTOEXEC.BAT, CONFIG.SYS, and SYSTEM.INI files; they might give clues as to which settings are already in use.
Turn off the machine and remove the cover. Be sure to take all appropriate measures for protection against electrostatic discharge (ESD).
Set the NICs jumpers or DIPP (dual inline package) switches as necessary and insert the card.
Turn on the machine and run the setup utility provided by the manufacturer. If you are using Windows 95, Windows 98, or Windows 2000, and the NIC is not Plug and Play, you can use the Add New Hardware wizard in the Control Panel to install the drivers and set up the card. (Remember to document all settings.)
A network protocol is a set of rules that govern the way computers communicate over a network. In order for computers using different software to communicate, they must follow the same set of networking rules and agreements, called protocols. A protocol is like a language; unless both computers are speaking and listening in the same language, no communication will take place.
Networking protocols are grouped according to their functions, such as sending and receiving messages from the NIC, or talking to the computer hardware and making it possible for applications to function in a network. Early computer networks had manufacturer-unique inflexible hardware and strict protocols. Today's protocols are designed to be open, which means they are not vendor-, hardware-, or software-specific. Protocols are generically referred to as protocol families or protocol suites because they tend to come in groups (usually originating from specific vendors).
To see the network components, including protocols, which are associated with a network connection, open the Network Connections folder, right click, the connection, and select Properties.
Here are the components that XP/2000 installs by default:
To see the settings for a particular protocol, click the protocol and then click Properties.
By default, XP/2000 configures TCP/IP to obtain an IP address automatically. If there's a DHCP server on the network, it will assign the IP address and other TCP/IP settings to the connection. Otherwise, Windows XP/2000 will use Automatic Private IP Addressing to assign an IP address to the connection.
This default configuration should work, unchanged, to connect a Windows XP computer to a network that uses TCP/IP for File and Printer Sharing in these common configurations:
One computer on the network is running Internet sharing software, such as Internet Connection Sharing, and provides a DHCP server for assigning TCP/IP settings to the other computers.
A hardware router provides shared Internet access and a DHCP server.
All computers run either Windows 98, 98SE, Me, 2000, or XP, with no DHCP server. The computers can use Automatic Private IP Addressing to assign themselves compatible IP addresses.
Using an Internet sharing program or a hardware router protects the local area network from access by other Internet users, so it's safe to use TCP/IP for File and Printer Sharing on the LAN. The computers have private IP addresses that aren't accessible from the Internet. No other protocol is needed.
If your network uses static IP addresses, click Use the following IP address and enter the configuration information. For example, here are possible settings for a network that uses a proxy server at IP address 192.168.1.1 for Internet access.