TYPES OF NETWORKS AND TOPOLOGY
Each type of network requires cabling, network equipment, file servers, workstations, software and training. You can combine these elements in various ways to create a network to accommodate the needs and resources of a particular organization. Some types of networks have low start-up costs, but are expensive to maintain or upgrade. Others are more expensive to set up, but are very easy to maintain and offer simple upgrade paths.
One of the clearest distinctions among types of networks is the network topology. The topology is the physical layout of a network combined with its logical characteristics. The physical layout is like an overhead picture or map of how the cabling is laid in the office, building or campus. The logical side of the network is the way the signal is transferred from point-to-point along the cable. The network layout may be decentralized, with cable running between all stations on the network, or the layout may be centralized, with each station physically connected to a central device that dispatches packets from workstation to workstation. Centralized layouts are like a star with workstations as its points. Decentralized layouts resemble a team of mountain climbers, with each climber at a different location on the mountain but all joined by a long rope. The logical side of a topology consists of the path taken by a packet as it travels through the network. We already mentioned in Chapter 1 that there are three main topologies: bus topology, ring topology and star topology.
In selecting a topology you should match the type of network with its intended purpose in an organization. For example, some organizations make more intense use of their network than others. The number and kind of software applications used by an organization influence the number and frequency of packets to be transmitted, which is known as network traffic. If network users are primarily accessing word-processing software, the network traffic will be relatively low, and most of the work will be performed at workstations, rather than on the network. Client/server applications generate a medium to high level of network traffic, depending on the client/server software design. Networks on which there is frequent exchange of database information, such as Microsoft SQL server or Oracle database files. have medium to high network traffic. Scientific programs and publications software generate high levels of traffic because they use extremely large data files. Also, graphics intensive applications such as streaming multimedia or desktop conferencing can produce very high levels of network traffic.
The impact of hosts and servers on a network is closely linked to the type of software applications that are used. For example, frequently accessing a database server to generate reports of financial and sales figures is likely to generate more network traffic than occasionally accessing a file server containing business correspondence or templates for letters.
You should consider whether other networks will be connected to a network, when deciding which topology to use. The network topology for a small business that will never use more than four computers will be different from the topology required by an industrial campus that connects via a WAN to other industrial sites.
The small business will probably not connect to additional networks, except for perhaps, an outside connection to the Internet. The industrial campus will consist of several interconnected networks, perhaps including a network to control machines in the plant, a network for the business systems, a network for the research scientists, and an extended WAN to other locations. Some topologies permit better network interconnectivity than other topologies.
Heavily trafficked networks need high-speed data transmission capabilities. The network speed is important to the productivity of the users. High-speed capability is particularly important when images, graphics and other large files need to be transported over long distances or onto WANs.
Security, i.e. the protection of data so that only authorized persons have access, is another issue that influences network design. A secure network uses network devices, passwords, control software and other techniques to restrict access to information and resources. It may also use data encryption. which encodes packets and allows only authorized computers to decode them. High-security networks use fiber­optic cable, which minimizes the risk of data interception by unauthorized users. Another security measure is to place network devices and servers in restricted locations, such as in computer rooms and wiring closets.
Network topology directly affects the network's potential for growth. After you install a network, you will probably need to add more users. perhaps in the same office or perhaps in other offices or on other floors. It is also likely that you will need to connect a LAN to a WAN for long-distance information access.
When a new LAN is installed, there are Several factors that affect its design, including the following:

LAN topology needs to be periodically analyzed, to ensure that the LAN meets the existing demands and also has the ability to address faster growth. The following questions should be considered:
Is there a significant increase in the number of network users?
Is there a change in the types of user workstations and applications? Are users demanding additional network service?
Is the network relatively easy to manage?
Are there demands for greater network reliability and additional redundancy? Can the existing network equipment be upgraded, or is it out of date?

LAN TRANSMISSION EQUIPMENT
LAN transmission equipment is used to connect devices on a single network, to create and connect multiple networks or sub-networks, and to set up a campus enterprise. The transmission equipment used on LANs can be deployed to connect a single node, or to interconnect multiple nodes and include the following:

• Network Interface Card (NIC) Repeaters
• Hubs
• Bridges
• Routers
• Brouters
• Switches Gateways.

Network Interface Card (NIC)
NIC is used to enable a network device, such as a computer or other network equipment, to connect to the network. NICs are designed to match particular network transport methods, computer bus types, and network media. The network connection requires four components:
An appropriate connector for the network medium A transceiver.A controller to support the Media Access Control (MAC) data link protocol Protocol control firmware.
The connector and its associated circuits are designed for a specific type of medium, for example, coax, twisted pair or optical fibre. Some NICs are made with multiple connectors so they can be used with different media. These combination NICs most commonly combine coax and twisted pair compatibility. When a combination NIC is used, it comes with software drivers or firmware to match the media options. Firmware is software that is stored on a chip, for example a ROM. Also, some NIC drivers are able to detect the medium attached to the NIC and then automatically set up the correct driver for the medium.

The cable connector is attached to the transceiver, which may be external to the NIC or built into it. For most computers, servers and network equipment, the transceiver is built into the interface card. In some cases, most often in older network equipment, the transceiver is external to the card and a transceiver drop cable is used to connect the transceiver to the card. (The transceiver drop cable is only necessary when the transceiver is external to the NIC and should not be used when the transceiver is built into the NIC.)
The MAC controller (MAC: Medium Access Control) unit and the firmware work together to correctly encapsulate source and destination address information, the data to be transported and the CRC (Cyclic Redundancy Check) information into the service data unit. The controller unit also ensures that the NIC waits 9.6 micro seconds to provide a small pause or "ideal period" between receipt of a frame and its subsequent transmission to another frame. The ideal period allows the WC to properly switch between receive and transmit modes. The MAC controller and firmware are customized for the particular type of network transport, which can be any one of the following:
• Ethernet
• Fast Ethernet
•  Gigabit Ethernet
• Token Ring
• Fast Token Ring
•   Fast Distributed Data Interface (FDDI)
•  Asynchronous Transfer Mode (ATM).
Some NICs combine network transport options, particularly for Ethernet and fast Ethernet, so that it is easy to upgrade a network for high speed communication. Also, many NICs are able to handle both half-duplex and full-duplex transmission.

BACK