CISCO4

The Internet is a worldwide collection of computer networks, cooperating with each other to exchange information using common standards. Through telephone wires, fiber optic cables, wireless transmissions and satellite links, Internet users can exchange information in a variety of forms.

The Internet is a network of networks that connects users in every country in the world. There are currently over one billion Internet users worldwide.

Any home, business or organization that wants to connect to the Internet must use an Internet Service Provider (ISP). An ISP is a company that provides the connections and support to access the Internet. It can also provide additional services such as Email and web hosting.

ISPs are essential to gaining access to the Internet. No one gets on the Internet without a host computer, and no one gets on the Internet without going through an ISP.

ISPs range in size from small to very large and differ in terms of the area they service. ISPs may provide limited services to a small geographical area or can have a wide variety of services and support entire countries with millions of customers. ISPs also differ in the types of connection technologies and speeds they offer. Examples of well known ISPs include AOL, EarthLink, and Roadrunner.

Do you have Internet access? Who is your ISP?

Individual computers and local networks connect to the ISP at a Point of Presence (POP). A POP is the connection point between the ISP's network and the particular geographical region that the POP is servicing.

An ISP may have many POPs depending on its size and the area it services. Within an ISP, a network of high-speed routers and switches move data between the various POPs. Multiple links interconnect the POPs to provide alternate routes for the data should one link fail or become overloaded with traffic and congested.

ISPs connect to other ISPs in order to send information beyond the boundaries of their own network. The Internet is made up of very high-speed data links that interconnect ISP POPs and ISPs to each other. These interconnections are part of the very large, high capacity network known as the Internet Backbone.

Connecting to the ISP at the POP provides users with access to the ISP's services and the Internet.

In a major city there are typically more choices for ISPs and more connection options than in a rural area. For example, cable Internet access is only available in certain metropolitan areas where cable TV service is available. Remote areas may only have access via dial-up or satellite.

Each Internet access technology uses a network access device, such as a modem, in order to connect to the ISP. It may be built in to your computer or may be provided by the ISP.

The simplest arrangement is a modem that provides a direct connection between a computer and the ISP. However, if multiple computers connect through a single ISP connection, you will need additional networking devices. This includes a switch to connect multiple hosts on a local network, and a router to move packets from your local network to the ISP network. A home networking device, such as an integrated router, can provide these functions, as well as wireless capability, in a single package.

Most of the technologies shown are used for both home and small business. Leased lines are typically used for business and large organizations, but can be used to provide high speed connectivity in areas where cable or DSL are not available.

Home service is normally less expensive than business services, and generally provides scaled-down services such as slower connection speed, reduced web space storage, and fewer email accounts. A typical home account may include a minimum of five email addresses with additional addresses being available for a fee.

Business class service is more expensive but provides faster connection speeds and additional web space and email accounts. A business class service may include twenty, fifty or more email addresses. Business service also includes agreements between the ISP and the customer specifying items such as network availability and service response time. These are known as Service Level Agreements (SLAs).

Asymmetric:

Most commonly used for the home.

Download speeds are faster than upload speeds.

Necessary for users that download significantly more than upload.

Most Internet users, especially those who use graphics or multimedia intensive web data, need lots of download bandwidth.

Symmetric:

Most commonly used for business or individuals hosting servers on the Internet.

Used when necessary to upload large amounts of traffic such as intensive graphics, multimedia, or video.

It can carry large amounts of data in both directions at equal rates.

For hosts to communicate on the Internet, they must be running Internet Protocol (IP) software. The IP protocol is one of a group of protocols that are collectively referred to as TCP/IP (Transmission Control Protocol / Internet Protocol). The Internet Protocol (IP) uses packets to carry data. Whether you are playing an Internet video game, chatting with a friend, sending email or searching the Web, the information you are sending or receiving is carried in the form of IP packets.

Each IP packet must contain a valid source and destination IP address. Without valid address information, packets sent will not reach the destination host. Return packets will not make it back to the original source.

IP defines the structure of the source and destination IP addresses. It specifies how these addresses are used in routing of packets from one host or network to another.

All protocols that operate on the Internet, including IP, are defined in numbered standards documents called RFCs (Request for Comments).

An IP packet has a header at the beginning which contains the source and destination IP addresses. It also contains control information that describes the packet to network devices, such as routers, it passes through and also helps to control its behavior on the network. The IP packet is sometimes referred to as a datagram.

IP addresses must be unique on the Internet. There are organizations responsible for controlling the distribution of IP addresses so that there is no duplication. ISPs obtain blocks of IP addresses from a local, national or regional Internet registry (RIR). It is the responsibly of the ISPs to manage these addresses and assign them to end users.

Before being sent on the Internet, messages are divided into packets. IP packet size is between 64 to 1500 bytes for Ethernet networks, and contains mostly user data. Downloading a single 1 MB song would require over 600 packets of 1500 bytes. Each individual packet must have a source and destination IP address.

When a packet is sent across the Internet, the ISP determines whether the packet is destined for a local service located on the ISP network, or a remote service located on a different network.

Every ISP has a control facility for their network, known as the Network Operations Center (NOC). The NOC usually controls traffic flow and houses services such as email and web hosting. The NOC may be located at one of the POPs or at a completely separate facility within the ISP network. Packets looking for local services are usually forwarded to the NOC and never leave the ISP network.

Routers in each of the ISP POPs use the destination address of the IP packets to choose the best path through the Internet. The packets you send to the ISP POP are forwarded by routers through the ISP's network and then through the networks of other ISPs. They pass from router to router until they reach their final destination.

The ping utility tests end-to-end connectivity between source and destination. It measures the time that it takes test packets to make a round trip from the source to the destination and whether the transmission is successful. However, if the packet does not reach the destination, or if delays are encountered along the way, there is no way to determine where the problem is located.

The traceroute utility traces the route from source to destination. Each router through which the packets travel is referred to as a hop. Traceroute displays each hop along the way and the time it takes for each one. If a problem occurs, the display of the time and the route that the packet traveled can help to determine where the packet was lost or delayed. The traceroute utility is called tracert in the Windows environment.

There are also a number of visual traceroute programs that can provide a graphical display of the route that a packet takes.

Therefore, in network diagrams a cloud is often used to represent the Internet or any other complex network, without showing the details of the connections. The cloud allows for simple diagrams that focus on source and destination only, even though there may be many devices linked in-between.

Devices that provide connectivity to end-users must match the technology used by the end-user to connect to the ISP. For example, if the end-user is using DSL technology to connect, the ISP must have a DSL Access Multiplexer (DSLAM) to accept these connections. For cable modems to connect, the ISP must have a Cable Modem Termination System (CMTS). Some ISPs still accept analog calls through modems and have banks of modems to support these users. ISPs that provide wireless access have wireless bridging equipment.

The ISP must also be able to connect with and transfer data with other ISPs. A variety of technologies are used to accomplish this, each requiring specialized equipment and configurations in order to function.

Networking devices used by the ISP handle extremely large volumes of traffic very quickly. They must function at near 100% uptime since the failure of a key piece of equipment at an ISP can have disastrous effects on network traffic. For this reason, most of the equipment used by ISPs are high-end, high-speed devices with redundancy.

In contrast, network devices used in the home or small business environment are lower-end, lower-speed devices that are not capable of handling large volumes of traffic. Integrated routers can perform several functions, including: Wireless LAN access point, switching, routing, firewalls and various address functions. An integrated router may support some or all of these functions.

The home or small business network provides a limited number of services for relatively few users. Internet connectivity is purchased from an ISP. The volume of traffic is small, and no transport services are provided.

The ISP provides transport and other services to a large number of users. A number of different devices are required to accept input from end users. To participate in a transport network, they must be able to connect to other ISPs. They handle large volumes of traffic and require very reliable equipment in order to handle the load.

The requirements are the same, but the scale of operation is different: at home, a single power outlet will suffice, whereas at an ISP the power requirements need to be planned out ahead of time and installed.

Environmental factors, such as heat and humidity, must also be considered when planning a network installation. However, because of the volume of equipment and the amount of power consumed in an ISP, high-end air conditioning units are necessary to maintain controlled temperatures. For the home/small business, ordinary air conditioning, heating, and humidity controls are usually sufficient.

Cable management is another area of concern for both the home/small business network and the ISP. Cables must be protected from physical damage and organized in a manner that will aid in the troubleshooting process. In small networks, there are only a few cables, but in ISP networks, thousands of cables must be managed. This can include not only copper data cables but also fiber optic and power cables.

All of these factors, namely power supply, environment and cable management, must be considered when setting up a network of any size. There is a big variation between size and therefore requirements for an ISP and a home network. Most networks fall somewhere between these two extremes.

A channel, or medium, provides a path over which the information is sent. In the networked world, the medium is usually some sort of physical cable. It may also be electromagnetic radiation, in the case of wireless networking. The connection between the source and destination may either be direct or indirect, and may span multiple media types.

Many different types of cables exist to interconnect the various devices in a NOC or local network.

There are two kinds of physical cable. Metal cables, usually copper, have electrical impulses applied to them to convey information. Fiber optic cables, made of glass or plastic, use flashes of light to convey information.

Twisted Pair

Modern Ethernet technology generally uses a type of copper cable known as twisted pair (TP) to interconnect devices. Because Ethernet is the foundation for most local networks, TP is the most commonly encountered type of network cabling.

Coaxial Cable

Coaxial cable is usually constructed of either copper or aluminum, and is used by cable television companies to provide service. It is also used for connecting the various components which make up satellite communication systems.

Fiber Optic

Fiber optic cables are made of glass or plastic. They have a very high bandwidth, which enables them to carry very large amounts of data. Fiber is used in backbone networks, large enterprise environments and large data centers. It is also used extensively by telephone companies.

Twisted pair cables consist of one or more pairs of insulated copper wires that are twisted together and housed in a protective jacket. Like all copper cables, twisted pair uses pulses of electricity to transmit data.

Data transmission is sensitive to interference or noise, which can reduce the data rate that a cable can provide. A twisted pair cable is susceptible to electromagnetic interference (EMI), a type of noise.

A source of interference, known as crosstalk, occurs when cables are bundled together for long lengths. The signal from one cable can leak out and enter adjacent cables.

When data transmission is corrupted due to interference such as crosstalk, the data must be retransmitted. This can degrade the data carrying capacity of the medium.

In twisted pair cabling, the number of twists per unit length affects the amount of resistance that the cable has to interference. Twisted pair cable suitable for carrying telephone traffic, referred to as CAT3, has 3-4 turns per foot making it less resistant. Cable suitable for data transmission, known as CAT5, has 3-4 turns per inch, making it more resistant to interference.

There are three types of twisted pair cable: unshielded twisted pair, shielded twisted pair, and screened twisted pair.

Unshielded twisted pair (UTP) is the most commonly encountered type of network cable in North America and many other areas. Shielded cables (ScTP and F-UTP) are used almost exclusively in European countries.

UTP cable is inexpensive, offers a high bandwidth, and is easy to install. This type of cable is used to connect workstations, hosts and network devices. It can come with many different numbers of pairs inside the jacket, but the most common number of pairs is four. Each pair is identified by a specific color code.

Many different categories of UTP cables have been developed over time. Each category of cable was developed to support a specific technology and most are no longer encountered in homes or offices. The cable types which are still commonly found include Categories 3, 5, 5e and 6. There are electrical environments in which EMI and RFI are so strong that shielding is a requirement to make communication possible, such as in a noisy factory. In this instance, it may be necessary to use a cable that contains shielding, such as Shielded twisted-pair (STP) and Screened twisted-pair (ScTP). Unfortunately both STP and ScTP are very expensive, not as flexible, and have additional requirements due to the shielding that make them difficult to work with.

All Categories of data grade UTP cable are traditionally terminated into an RJ-45 connector.

Like twisted pair, coaxial cable (or coax) also carries data in the form of electrical signals. It provides improved shielding compared to UTP, so has a lower signal-to-noise ratio and can therefore carry more data. It is often used to connect a TV set to the signal source, be it a cable TV outlet, satellite TV, or conventional antenna. It is also used at NOCs to connect to the cable modem termination system (CMTS) and to connect to some high-speed interfaces.

Although coax has improved data carrying characteristics, twisted pair cabling has replaced coax in local area networking uses. Among the reasons for the replacement is that - compared to UTP - coax is physically harder to install, more expensive, and harder to troubleshoot.

Unlike TP and coax, fiber optic cables transmit data using pulses of light. Although not normally found in home or small business environments, fiber optic cabling is widely used in enterprise environments and large data centers.

Fiber optic cable is constructed of either glass or plastic, neither of which conducts electricity. This means that it is immune to EMI and is suitable for installation in environments where interference is a problem.

In addition to its resistance to EMI, fiber optic cables support a large amount of bandwidth making them ideally suited for high-speed data backbones. Fiber optic backbones are found in many corporations and are also used to connect ISPs on the Internet.

Each fiber optic circuit is actually two fiber cables. One is used to transmit data; the other is used to receive data.

There are two forms of fiber optic cable: multimode and single mode.

Multimode

Of the two forms of fiber optic, multimode is the less expensive and more widely used. The light source that produces the pulses of light is usually an LED. It is referred to as multimode because there are multiple rays of light, each carrying data, being transmitted through the cable simultaneously. Each ray of light takes a separate path through the multimode core. Multimode fiber optical cables are generally suitable for links of up to 2000 meters. However, improvements in technology are continually improving this distance.

Single Mode

Single mode fiber optic cables are constructed in such a way that light can follow only a single path through the fiber. The light source for single mode fiber optic cables is usually a LED laser, which is significantly more expensive and intense than ordinary LEDs. Due to the intensity of the LED laser, much higher data rates and longer ranges can be obtained. Single mode fibers can transmit data for approximately 3000 meters and are used for backbone cabling including the interconnection of various NOCs. Again, improvements in technology are continually improving this distance.

Cabling is an integral part of building any network. When installing cable, it is important to follow cabling standards, which have been developed to ensure data networks operate to agreed levels of performance.

Cabling standards are a set of specifications for the installation and testing of cables. Standards specify types of cables to use in specific environments, conductor materials, pinouts, wire sizes, shielding, cable lengths, connector types and performance limits.

There are many different organizations involved in the creation of cabling standards. While some of these organizations have only local jurisdiction many offer standards that are adopted around the world.

Some of the organizations and the areas that they manage are seen in the graphic.

Twisted pair cable is most commonly used in network installations. The TIA/EIA organization defines two different patterns, or wiring schemes, called T568A and T568B. Each wiring scheme defines the pinout, or order of wire connections, on the end of the cable.

The two schemes are similar except two of the four pairs are reversed in the termination order. The graphic shows this color-coding and how the two pairs are reversed.

On a network installation, one of the two wiring schemes (T568A or T568B) should be chosen and followed. It is important that the same wiring scheme is used for every termination in that project. If working on an existing network, use the wiring scheme already employed

Specific pins on the connector are associated with a transmit function and a receive function. The transmit pin versus the receive pin is determined based on the device.

Two devices directly connected and using different pins for transmit and receive are known as unlike devices. They require a straight-through cable to exchange data. Devices that are directly connected and use the same pins for transmit and receive, are known as like devices. They require the use of a crossover cable to exchange data.

Unlike Devices

The pins on the RJ-45 data connector of a PC have pins 1 and 2 as transmit and pins 3 and 6 as receive. The pins on the data connector of a switch have pins 1 and 2 as receive and pins 3 and 6 as transmit. The pins used for transmit on the PC correspond to those used for receive on the switch. Therefore, a straight-through cable is necessary.

The wire connected to pin 1 (transmit pin) on the PC on one end of the cable, is connected to pin 1 (receive pin) on the switch on the other end of the cable.

Other examples of unlike devices that require a straight-through cable include:

Switch port to router port

Hub port to PC

f a PC is directly connected to another PC, pins 1 and 2 on both devices are transmit pins and pins 3 and 6 are receive pins.

A crossover cable would ensure that the green wire connected to pins 1 and 2 (transmit pins) on one PC connect to pins 3 and 6 (receive pins) on the other PC.

If a straight-through cable were used, the wire connected to pin 1, the transmit pin, on PC1 would be connected to pin 1, the transmit pin, on PC2. It is not possible to receive data on a transmit pin.

Other examples of like devices that require a crossover cable include:

Switch port to switch port

Switch port to hub port

Hub port to hub port

Router port to router port

PC to router port

PC to PC

If the incorrect cable type is used, the connection between network devices will not function.

Some devices can automatically sense which pins are used for transmit and receive and will adjust their internal connections accordingly.

UTP and STP cable is usually terminated into an RJ-45 connector.

The RJ-45 connector is considered a male component, which is crimped to the end of the cable. When a male connector is viewed from the front with the metal contacts facing up, the pin locations are numbered from 8 on the left to 1 on the right.

The jack is considered the female component and is located in networking devices, wall outlets, or patch panels. The RJ-45 connector on the wire plugs into the jack.

Cables can be purchased that are pre-terminated with RJ-45 connectors. They can also be manually terminated, onsite, using a crimping tool. When manually terminating UTP cable into an RJ-45 connector, untwist only a small amount of wire to minimize crosstalk. Also be sure that the wires are pushed all the way into the end of the connector and that the RJ-45 connector is crimped onto the wire jacket. This ensures good electrical contact and provides strength to the wire connection.

In a NOC, network devices are usually connected to patch panels. Patch panels act like switchboards that connect workstation cables to other devices. The use of patch panels enables the physical cabling of the network to be quickly rearranged as equipment is added or replaced. These patch panels use RJ-45 jacks for quick connection on the front, but require the cables to be punched down on the reverse side of the RJ-45 jack.

Patch panels are no longer confined to enterprise network installations. They can be found in many small businesses and even homes where they provide a central connection point for data, telephone and even audio systems.

The RJ-45 jack has eight conductors, and is wired according to either T568A or T568B. At the patch panel a device known as a punchdown tool is required to push the wires into the connector. The wires should be matched up to the appropriate insulation displacement connector (IDC) by color before punching them down. The punchdown tool also cuts off any excess wire.

A punchdown tool is not required to terminate most wall jacks. To terminate these connectors the cables are untwisted and placed into the appropriate IDC. Placing the cap on the jack pushes the cables into the IDC and cuts through the insulation on the wires. Most of these connectors then require the installer to manually trim away excess cable.

In all cases, untwisting more cable than is necessary will increase the amount of crosstalk and degrade overall network performance.

When a new or repaired cable run is terminated, it is important to verify that the cable operates correctly and meets connectivity standards. This can be done through a series of tests.

The first test is a visual inspection, which verifies that all wires are connected according to T568A or B.

In addition to visual examination, check the cable electrically in order to determine problems or flaws in a network cabling installation. The following are tools that can be used for cable diagnostics:

Cable testers

Cable certifiers

Multimeters

The cable tester is used to perform initial diagnostics. The first test usually is called a continuity test and it verifies that there is end-to-end connectivity. It can also detect common cabling faults such as opens and shorts.

An open circuit occurs when the wire is not properly pushed into the connector and there is no electrical contact. An open can also occur if there is a break in the wire.

A short occurs when the copper conductors touch each other. As the electric pulse travels down the wire, it will cross onto the touching wire. This creates an unintended path in the flow of the signal to its destination.

A cable tester can also create wire maps that will verify that the cable is terminated correctly. A wire map shows which wire pairs connect to which pins on the plugs and sockets. The wire map test verifies that all eight wires are connected to the correct pins and indicates if cabling faults are present such as split pairs or reversals.

If any of these faults are detected, the easiest way to correct them is to reterminate the cable.

Specialized cable testers provide additional information, such as the level of attenuation and crosstalk.

Attenuation

Attenuation, also commonly referred to as insertion loss, is a general term that refers to the reduction in the strength of a signal. Attenuation is a natural consequence of signal transmission over any medium. Attenuation limits the length of network cabling over which a message can be sent. For example, Ethernet cable has a distance limitation of 328 feet (100 meters) where as some types of fiber optic cable have a distance limitation of several miles (kilometers). A cable tester measures attenuation by injecting a signal in one end and then measuring its strength at the other end.

Crosstalk

Crosstalk is the leakage of signals between pairs. If this is measured near the transmitting end it is termed near-end crosstalk (NEXT). If measured at the receiving end of the cable it is termed far-end crosstalk (FEXT). Both forms of crosstalk degrade network performance and are often caused by untwisting too much cable when terminating. If high crosstalk values are detected, the best thing to do is check the cable terminations and re-terminate as necessary.

The following steps, called best practices, ensure that cable termination is successful.

1. It is important that the type of cables and components used on a network adhere to the standards required for that network. Modern converged networks carry voice, video and data traffic on the same wires; therefore the cables used on converged networks must be able to support all these applications.

2. Cable standards specify maximum lengths for different types of cables. Always adhere to the length restrictions for the type of cable being installed.

3. UTP, like all copper cable, is susceptible to EMI. It is important to install cable away from sources of interference such as high-voltage cables and fluorescent lighting. Televisions, computer monitors and microwaves are other possible sources of interference. In some environments it may be necessary to install data cables in conduit to protect them from EMI and RFI.

4. Improper termination and the use of low quality cables and connectors can degrade the signal carrying capacity of the cable. Always follow the rules for cable termination and test to verify that the termination has been done properly.

5. Test all cable installations to ensure proper connectivity and operation.

6. Label all cables as they are installed, and record the location of cables in network documentation.

Structured cabling is a method for creating an organized cabling system that can be easily understood by installers, network administrators, and any other technicians who deal with cables. One component of structured cabling is cable management.

Cable management serves multiple purposes. First, it presents a neat and organized system which aids in the isolation of cabling problems. Second, by following cable management best practices, the cables are protected from physical damage which greatly reduces the number of problems experienced.

Cables should be considered a long term investment. What may be sufficient now may not be in the near future. Always plan for the future by complying with all current standards. Remember that standards help to ensure that the cables will be able to deliver acceptable performance as the technology evolves.

It is important to observe cabling best practices in all environments. Strict adherence to these practices, in home and business environments, helps reduce the number of potential problems. It will save a great amount of time, money and frustration.