I don’t have a lot of Tech savvy friends and relatives, so whenever I tell then what I majored in school, I always end up cutting it down to…”I get computers to talk to each other and if they don’t; I figure out why not.” So here is exactly what a network is…
A network is a collection of electronic devices that are connected using a medium, or media, to transmit and receive data. The medium used is contingent on the size and goals of the owner of the network. The purpose of a network is to share resources and send/receive data.
For example, Tifa Institute of Networking has a campus with three learning buildings and two student residence buildings within . The President of the school wants to have Wi-Fi (short for Wireless Fidelity) throughout the campus for students and classroom/lecture hall use and a second, separate wired network for faculty, staff and administrators.
Now in this case, the network would be called a CAN or Campus Area Network, because there are multiple buildings in close proximity that are on the same network.
Now, don’t think that this is the only type of network, because it is most certainly not. Networks come in all shapes (topologies) and sizes. Think about it… you are connected to a network right now. You may be a connected to a cellular network, a home Wi-Fi network, or a public wired network or Wi-Fi network accessing the Internet.
The purpose of a network is to share resources and transit/receive data. Network resources consist of:
- Hardware: routers, switches, wireless access points, servers, and printers.
- Applications: E-mail, Web, instant messaging, file sharing, video conferencing
- Data: documents, images, etc
- Storage: for data back-up and retrieval
- Communication: IP phones, PBX, for instance.
This discussion will get deeper in later posts but for now, I’ll just give you the cliff notes. Network components are essentially the tools used within a network to carry out its functions. As I type and as you read, we are taking advantage of a few network components. Here are the network components, or at least most of them:
- client: your computer or device. An end-user device that sends requests to a server.
- server: responds to client devices with the resources requested or an error message for various reasons.
- hub: has ports to which client devices connect to send and receive data; however, it doesn’t think, it just pushes signals through whether they make it to their destination or not. As a result, collisions and looping occur causing performance lag on the network.
- bridge: connects separate networks.
- switch: prevents collisions by separating collision domains and prevents loops by keeping a table that maps its ports and activity… switches do lots, we’ll dive in later…
- router: makes it possible for networks to send and receive messages out to other networks.
- media: the cords. the links that are used to travel the electromagnetic or light pulses to and from routers.
Networks are categorized by distance and size. So here’s the breakdown, largest to smallest:
- Wide Area Network (WAN), connects networks across states and countries
- Metropolitan Area Network (MAN), connects networks within cities and states.
- Campus Area Networks (CAN), connects networks within close proximity, about a few miles, to each other.
- Local Area Network (LAN), connects a network within a single location, can be one building with multiple floors, or one floor in a building.
- Personal Area Network (PAN) connects a small network for personal use, usually a home network or SOHO (small office/ home office).
There are two topologies: physical and logical. A physical topology is the manner in which a network is physically connected: star, ring (or dual-ring), bus, mesh (full- or partial-).
- star: there is a central device, usually a switch, to which the endpoints connect. This also presents a single point of failure (thumbs down).
- ring: each device is connected to the device to its right, like kids holding hands in a circle. The disadvantage is that if a link breaks or an endpoint goes down, so does the network.
- dual ring: the same as above but with redundancy, to try to mitigate the risk of complete network failure on the occasion that a device or link fails, and tons of luck troubleshooting that nightmare, LOL.
- bus: the endpoints are connected to a backbone cable using a network tap, also formally known as a T-connector. The bad news: the cable is a single point of failure.
- mesh: every device has a direct connection with every other device. If you enjoy tedium, this ones for you. Oh and it’s very expensive because of all the wiring. On the bright side, it is the most stable, not easily scalable, but stable.
A logical topology is the manner of which the data travels. So please understand that the physical and logical topology DO NOT have to be the same, and probably will not be the same.
Peer-to-peer and Client-server…
With client-server network, you get better performance because the baby is getting everything it needs from Momma.
Client: “Mommy I wanna go to YouTube!”
Server: “What’s that sweetie? HTTP request? ok there you have it, Uncle web server is right here to help.”
Ok so maybe the mini-skit was pretty bad, but I hope that bit of humor, or lack thereof, makes it a little bit easier to remember. When you’re in an office setting, or any environment where you have multiple devices that are sharing resources, this is best option.
In a peer-to-peer network, as the name implies. Peers give resources to and receive resources from each other. Its like ebay or letgo for networks. I’m probably telling my age, but… its like Kazaa or Morpheus, I am not admitting to anything, it’s just an example. I don’t know about you, but I remember things best when I have a real example for the information I am trying to learn.