Understanding networking

As a manager, it is not your responsibility to know how to configure a router and make things work in the network.   The best way that you should consider networking is the “black box theory”.  You really don’t care how the individual parts work.  You need to know what they are capable of.  Believe it or not, networking is really simple.

At the simplest form, a network is a few computers that are connected by a wire to a network device that shares the information to each computer.   A network is similar to a big post office that is sharing information packets electronically.  The computers each have a unique name that helps the network devices know what information goes to what computer.

The internet is an IP-based network.  IP stands for Internet Protocol.  Easy, huh?  The Transmission Control Protocol is the way that the computers break up large data chunks to send across the internet.  Stick the two together and you’ll get the commonly referred to TCP/IP.  There are other forms of message handling—such as User Datagram Protocol (UDP)—to move information across the internet.  You don’t need to know how these work or move information.  Just know that IP is the backbone of the internet.

Any data that you can turn into an IP packet can travel over an IP network; that data can also travel local networks and the internet.  When a phone converts voice to an IP packet, it is called a Voice over IP phone (VoIP) meaning that it can send your phone call over the same network as email, web browsing, and everything else.

Blah-blah over IP is nothing fancy.  That means that someone has designed a network device (or interface) that translates information from a source to an IP-packet, and back.  You’ll hear about Radio-over-IP, Video-over-IP, Computer-over-IP, and just about everything else.

Data standards are really important in this area.  When each vendor comes up with their own way of doing ____-over-IP, then it is likely that vendors will not be compatible unless they use a standard.  While there are organizations that state standards, a standards true usefulness is proven by if and how people use it.

The International Telecommunications Union (ITU) has a series of standards for videoconferencing, including ITU H.320 and H.264.  When Cisco Telepresence was released, it was designed to bring a meeting room presence to teleconferencing.  Part of the design was full size displays that blended both conference rooms.  It was not compatible with any other video conferencing systems.  The Cisco sales rep explained to me that their product would look poor if it was used with lower quality non-telepresence systems, so the decision was made to be a non-standard data packet.  The problem with this is that it would require companies to invest in two separate video conferencing systems.  More recent advances have allowed some mixed use of video conferencing systems.

Now that we’ve talked a bit about what can go across the network, let us turn back to the network itself.

There are many different formats of networks.  A quick internet search on “network topology” will show the different forms.  Each has an advantage and a disadvantage.  For this course, the focus will be on a tree topology.  An internet connection enters a site through one point.  Switches and routers are used to split that internet connection to all the individual computers.

A demarc (short for demarcation) point is where a utility enters a building.  It is also the point that separates ownership between the utility company and the building owner.  The electrical demarc in a residential home is commonly the electric meter.  The power company will handle everything up to and including the meter.  The home owner handles everything from the meter to the power outlets.

A telephone demarc is located at the telephone network interface.  The network demarc is located at the network interface device (aka smartjack).  These can be located anywhere in a building, but I’ve found that most wireline utilities come in together.  These can be copper wire, fiber optic or some other type of cable.

The demarc is the head of the network for that site.  In a tree topology, this is where the site’s primary router would be located.  A router is a network device that moves data packets between two different networks.  Here, the router is directing the packets, only passing those that need to travel on the other network.  It is ideal for separating two networks to reduce congestion by keeping local data within the local network.  A primary router, sometimes called a site’s core router, is the one that controls the other routers and is mission-critical for the site to be connected.

Routers are the major component that give a network flexibility.  Professional (non-consumer) grade routers allow for the installation of modules, both physical and logical.  These modules connect the router to different devices.  These modules commonly allow a router to connect to a wireline (T1, T3, etc) circuit, a wireless (wifi, cell) circuit, or a different cabling (twisted pair, coax).  These modules can also be used to connect a router to a phone system, radio system, video system and so on.

Other network devices used to spread network segments out from the router include switches and hubs.  Switches can have different interfaces and be used to connect different network types.  This is handy in older buildings where you may need to use an existing style network and will overlay it with a different type of cabling or connections.  Hubs are almost non-intelligent splitters that just provide more ports.

The Warriors of the Net video provides an entertaining explanation of the different components.  Again, from a manager’s perspective, you do not need to get very technical with the network components.