Essential Network Concepts Part 1
Welcome to article number 4 in my Cisco CCNA in 15 minutes a week series. This week’s article begins the move beyond network models and into actual networking concepts. Prior to getting into the details of Cisco devices and their configuration, you’ll need a solid fundamental understanding of not only network concepts, but also technologies, protocols, and equipment. This article will be the first of a few that will begin covering some of these critical areas; this particular article provides an overview of basic network equipment.
The material to be covered in the article includes an overview of:
– Repeaters and hubs
– Bridges and Switches
In order to gain a perspective on networking, you’ll first need to understand the purpose and functions of different network devices. While you’ll certainly be familiar with much (if not all) of the equipment, you will need to understand both how they relate to network communication and the OSI model. The standard equipment that we’ll look at first includes repeaters, hubs, bridges, switches, and routers.
Devices at the OSI Physical Layer
A repeater is a network device with a rather singular purpose. Though you probably won’t see a repeater hanging around a server room to often any more (at least not Ethernet ones), their main function is to regenerate a single to allow it to travel greater distances along a network. As a signal travels along a network, it slowly begins to lose strength, especially after long distances. The simple purpose of a repeater is to amplify or regenerate that signal, allow it to carry on along its path. Note that a repeater is a Physical layer device, and doesn’t actually have any intelligence built in. It simple regenerates a signal – nothing more and nothing less. In the past, you might commonly find these on coaxial Ethernet networks, a small box (powered) that allowed the network to span greater distances.
A hub is another Physical layer device that you are more than likely familiar with. Much like a repeater, a hub also has no real intelligence – if this idea confuses you, just remember that a hub doesn’t make any decisions. What a hub does provide is connectivity, allowing attached devices a path between which they can communicate. Beyond that, the hub also acts a multiport repeater, regenerating signals as they pass between connected devices. When one device connected to a hub puts traffic onto the network, all attached devices receive it. This inefficiency leads to collisions, which we’ll describe in more detail as we move through the series.
Devices at the OSI Data Link Layer
Switches and bridges are examples of devices that can be found at the OSI Data Link layer. The key thing to remember is that these devices do have a degree of intelligence, in that they pay attention to the source and destination MAC addresses of frames. While a switch might often be mistaken for a hub, it is worth recognizing that when a frame is received by a switch (from an attached computer, for example), it takes a look at something referred to as its MAC address table, and figures out on which port the destination MAC address can be found. In cases where the MAC address is not a broadcast (or multicast), the frame is only forwarded out the necessary port, and is never seen by any other attached devices.
As such, what a switch does is often referred to as segmentation. It segments the network into collision domains (groups of computers between which collisions can occur) at the port level. Now, it’s worth recognizing that someone may have plugged a hub into a switch port, and attached many computers to that hub. In those cases, the port still represents a collision domain – there may be collisions among the computers plugged into the hub, but not between a computer plugged into that hub and one plugged into another switch port. To that end, switches provide higher performance than hubs by reducing (or eliminating) network collisions. Consider the diagram below.
A bridge provides the same basic function as a bridge, segmenting the network into different collision domains. However, bridges tend to older devices that you won’t often see anymore. Their bridging logic was usually implemented in software, whereas switches use what are referred to as Application Specific Integrated Circuits (ASICs). These provide very high throughput rates and make forwarding decisions quickly. In older environments, a bridge might only have two ports, basically separating the network into two collision domains. Think of a switch as a multiport bridge.
One last note on switches and bridges for the time being. Although a switch or bridge separates the network into different collision domains, it still passes all broadcasts and multicasts to all ports. This is because all of the systems will still be part of the same broadcast domain, a term I’ll describe shortly.
Devices at the OSI Network Layer
Devices at the Network layer are concerned with two main things – network addressing (such as IP addressing) and routing. As such, it is correct to assume that a router is a layer 3 device. To that end, it should also be noted that it is becoming increasingly common to come across what is referred to as a Layer 3 switch. In fact, this is a switch that not only does the switching function discussed in the previous section, but is also capable of routing between broadcast domains.
A broadcast domain is another simple but important topic. For all intents and purposes, a broadcast domain represents the systems to which a given broadcast will travel. For example, broadcasts do not pass routers by default – the router acts as a sort of demarcation point where broadcasts stop. This helps ensure that broadcasts from one part of the network don’t interfere with another. However, in order to communicate between broadcast domains, you require a suitable layer 3 device, such as a router. Just remember that routers break up broadcast domains. See the diagram below that shows both broadcast domains in green and collision domains in red on a simple routed network.
To that end, the router is also responsible for getting traffic from one broadcast domain to another as necessary. On an IP-based network, you provide computers with a default gateway, which specifies the address of their local router. When they need to communicate with a host on another broadcast domain they forward packets to the router, who then decides where they should be sent next based on entries in its routing table. Routers will always have at least two interfaces in a proper implementation, and often more. For example, a router might connect two Ethernet LANs, or an Ethernet LAN to a Token Ring LAN. In that way, a router can also be used to help connect networks that use different network technologies.
Dan DiNicolo is a technical trainer, consultant, author, and the managing editor of the free IT learning web site 2000Trainers.com. When he’s not busy traveling the world as an IT volunteer with organizations like Geekcorps, Dan makes his home in the snowy northern backwoods of Canada.