Introduction to Local and Wide-area Networks - Part 3

Introduction to Local and Wide-area Networks - Part 3
Page content

Local Area Networks

Before the introduction of networks, PCs were islands of information. To share information, a user copied information to a floppy disk and walked it across the office to the person requesting it. This was “affectionately” known as a sneaker net. In addition, a printer was attached to each PC, since there was no way for PC users to share computer devices. One of the biggest problems with this configuration was the lack of a central backup solution. Management had to rely on the diligence of each user to back up the information on her PC before she left for the day. I doubt if anyone works in an office like this today. But it’s nice to look back once in a while to see where we came from. Today’s network technology has changed the way business information is stored, shared, and processed.

Peer-to-peer networks Figure 1 shows a basic network configuration known as a peer-to-peer network. Each PC can use either its own internal disk storage, or it can share the storage of the other PC. Information is shared between the users of these two systems by each user granting file access to the other user. In addition, User A can use the printer connected to User B’s PC. This configuration works well when there are few devices connected to your network. Peer-to-peer networks solve the problem of sneaker net and unshared resources in very small offices, but they quickly become unmanageable. Information might still be distributed across multiple systems, and a central backup process is typically not supported. I don’t recommend this configuration for any size office.

Hub-centric networks

Figure 2 is a much better solution for connecting a small number of devices to a network. It depicts a network connected with a hub. This network configuration is a single network segment, a single collision domain, and a single broadcast domain. Network devices are normally connected to the hub with TP cable.

All packets transmitted by the connected devices travel through the cables to the hub. Since all ports in a hub are effectively connected by a shared path (see Figure 3), all ports receive all transmitted packets. This results in each connected device seeing all packets, even those intended for other devices. Although a device won’t process a packet addressed to another device, each NIC must examine the destination MAC address in each packet it sees to determine if the packet is intended for the device in which it’s installed. The receipt of a large number of packets intended for other destinations can result in a very busy NIC. If a NIC is busy, performance suffers.

Another important difference between the network in Figure 1 and the network in Figure 2 is the presence of a file server. A file server is used to provide central storage of flat file information. Ensuring critical information is backed up regularly is much easier. A file server can also double as a print server. All network PCs can use the server as a means to spool and print information to the shared network-attached laser printer.

This is a good time to discuss how a device obtains the MAC address of a device on the same network segment with which it wants to communicate. Figure 4 is our hub-centric network with device IP addresses shown. The subnet mask for this network is This means all the bits in the first three octets (counted from the left) are part of the network address. So the network address is The fourth octet in each IP address is the unique device or host address.

In a Microsoft Windows environment, a list of network resources and their IP addresses is available to each connected device. WS 1 looks up SVR 1 in the list and retrieves the server’s IP address.

Examining the IP address, WS 1 determines that the network address of SVR 1 is the same as its own network address. In other words, they are both in the same network segment. WS 1 checks the list of IP/MAC address pairs, called the Address Resolution Protocol (ARP) cache, it has incrementally constructed in its memory. The ARP cache contains addresses for all the devices with which WS 1 has recently communicated. In our example, SVR 1 is not contained in the list.

WS 1 sends a broadcast packet asking the device with IP address to respond with its MAC address. Since this is a broadcast packet, all devices on the network, including the printer, process the packet. However, only SVR 1 responds with a MAC address.

When WS 1 receives SVR 1’s MAC address, it stores it in its ARP cache for future use. It then creates an Ethernet packet with SVR 1’s MAC address as the Destination Address and its own MAC address as the Source Address. WS 1 places the packet on the network.

Every NIC in the network segment sees the packet and evaluates the Destination Address. Only SVR 1 will process the packet. If a response is required, SVR 1 uses WS 1’s MAC address, included in the original packet, to return a packet.

If SVR 1 is in a different network segment with a different network address, additional steps are necessary. We cover these additional steps later in this chapter.

Since this is a collision domain, all connected devices are competing for the same bandwidth. In a small network this is usually not an issue. In a large network, this can cause performance problems. One way to improve performance is to separate your network into multiple collision domains with a switch. Switched networks are the topic of Part 4.

Go to Part 4...

See other articles in this series…

Tables and Figures (Hover for caption, click to enlarge)

Figure 2: Hub-centric Network

Figure 3: Inside a Hub

Figure 4: Hub-centric Network with IP Addresses

Tables and Figures (Hover for caption, click to enlarge)

Collision - When two devices in the same Ethernet network segment attempt to send a packet at the same time, a collision occurs. Each device waits a random period before attempting to transmit again.

Collision Domain - A network segment in which the packets of two or more network devices may collide when transmitted at the same time.

File Server - A file server normally stores flat files rather than large production databases. User home directories are normally located here. A File server can also provide network printing services. A server that performs both functions is called a file and print server. A print server allows multiple users on a network to share a printer.

Flat File - In general, a flat file is a non-database file. We define a database as a production storage program such as Oracle or Microsoft SQL Server. Flat files include word processing documents, spreadsheet files, graphics, etc.

Local Area Network - A local area network (LAN) is a collection of devices connected together, over high speed links, to share computing resources. A LAN is typically deployed within a single office, building, or campus.

Network-attached Laser Printer - A printer attached to a network via a network cable, with its own network address, is a network-attached printer. Using a print server, network users can share the printer.

This post is part of the series: Introduction to Local and Wide Area Networks

Understanding how networks work is an important first step in understanding information security. This series provides everything you need to know to get started.

  1. Introduction to Local and Wide-area Networks - Part 1
  2. Introduction to Local and Wide-area Networks - Part 2
  3. Introduction to Local and Wide-area Networks - Part 3
  4. Introduction to Local and Wide-area Networks - Part 4
  5. Introduction to Local and Wide-area Networks - Part 5
  6. Introduction to Local and Wide-area Networks - Part 6
  7. The Importance of Wireless Connectivity