Home > Articles > Networking

  • Print
  • + Share This
This chapter is from the book

The Network Architecture: Combining the Physical and Logical Components

When computers are connected, we must choose a network architecture, which is the combination of all the physical and logical components. The components are arranged (we hope) in such a way that they provide us with an efficient transport and storage system for our data. The network architecture we choose dictates the physical topology and the logical arrangements of the system. For example, if I say, “I’m building a Switched Ethernet network,” this statement implies the overall architecture of my future network. Let’s now examine these physical and logical components.

The Physical Network

The physical network is easy to understand because it’s usually visible. Mainly, it consists of hardware: the wiring, plugs such as computer ports, printers, mail servers, and other devices that process and store our data. The physical network also includes the important (read: vital) signals that represent the user data. Examples are voltage levels and light pulses to represent binary images of 1s and 0s—strung together in many combinations to describe our data.

I say “usually visible” because we can’t see wireless connections. Although more ethereal than copper wire connections, wireless connections are nonetheless physical, taking the form of electromagnetic radio waves.

Quite rare only a few years ago, wireless networks such as Wi-Fi are now common. If you have a broadband connection in your home, chances are good your computer is connected to your broadband hardware device with a wireless arrangement. How we explain the layout (also called a topology) of a wireless network is no different from that of a wire-based network.

Physical Layout—Network Topologies

As mentioned, the physical aspect of the network consists of the components that support the physical connection between computers. In today’s networks, four topologies are employed: (a) star, (b) ring, (c) bus, and (d) cell. They are depicted in Figure 1.1.

Figure 1.1

Figure 1.1 Network topologies: (a) Star topology, (b) Ring topology, (c) Bus topology, (d) Cellular topology

  • Star—The star topology employs a central connection point, called a router, hub, bridge, or switch. The computers on the network radiate out from this point, as seen in Figure 1.1(a). The job of the central point is to switch (relay) the users’ data between user machines and perhaps other central connection points. The terms router, hub, bridge, or switch are used interchangeably by some people. Generally, the terms hub and bridge are associated with devices of a somewhat limited capacity. The term switch has historically been associated with telephone networks (with the exception of the 1970’s computer network message switches and 1980’s packet switches). The term router found its way into the industry in the 1980s and is now used more frequently than the other terms. Whatever we call these machines, they manage traffic on the network and relay this traffic back and forth between our computers.
  • Ring—The ring topology, shown in Figure 1.1(b), connects the computers through a wire or cable. As the data (usually called a packet) travels around the ring, each computer examines a destination address in the packet header (similar in concept to a postal envelope’s “to” address) and copies the data if the computer’s address matches its address. Otherwise, the computer simply passes the packet back onto the ring to the next computer (often called the next node). When the packet arrives at the originating node, it removes the packet from the ring by not passing it on.

    The ring topology is the first example of a broadcast network: Nodes in the network receive all traffic in the network. Whether a node chooses to accept the packet depends on the destination address in the packet header.

  • Bus—The bus topology is shown in Figure 1.1(c). It consists of a wire with taps along its length to which computers connect. It is also a broadcast network because all nodes receive the traffic. The sending node transmits the packet in both directions on the bus. The receiving nodes copy an image of the packet if the destination address matches the address of the node. The packet rapidly propagates through the bus, where it is then “terminated” at the two ends of the bus. As you may have surmised, packets traveling along this bus may interfere with each other if the nodes relay the packets onto the bus at about the same time. The bus topology handles this situation with a collision detection procedure. A node keeps sending until it detects its transmission has occurred without interference (by checking its own transmission).
  • Cellular—The cellular topology is employed in wireless networks, an arrangement shown in Figure 1.1(d). Cellular networks use broadcast protocols; all nodes (cellular phones) are capable of receiving transmissions on a control channel from a central site. A wireless control node (called the base station) uses this common channel to direct a node to lock onto a specific (user) channel for its connection. During the ongoing connection, the cell phone is simultaneously communicating with the base station with the control link and the user link.

The Logical Network

The previous section explained the physical layout of networks, such as the star topology. In explaining how packets of user traffic are moved across these topologies, we have also explained the logical aspects of a network. Again, the logical parts of computer networks entail the invocation of software to “propel” the packets across the physical media and to receive them at the other end.

Unlike the physical network, the logical network is not visible. It uses the physical network for transport of data. We defer describing the details of the logical network here, as it is described extensively in almost every subsequent hour.

  • + Share This
  • 🔖 Save To Your Account