Well, you made it through Part 3 of Security Sales & Integration’s four-part “Networked Video for D.U.M.I.E.S.” series, brought to you by Pelco. Congratulations!

While the last section was a lot of theory (see “Essentials of Network Design and Function” in the August issue), it provided the fundamentals for understanding the network as awhole, specifically how the hardware works to achieve yourgoal of putting video on the network.

The first two parts of “D.U.M.I.E.S.” dealt with getting video to the network. We talked about the quality of the cameras themselves and the installation (see “Essentials of Networked Cameras and Lenses” in the April issue). We also took a look at the different compression methods and the process required to turn analog video into digital (see “How to Beat the Bandwidth Blues” in the June issue).

In this installment, we will deal with the implementation of the network. We’ll look at the hardware you’ll run into and the technologies you will have to know to successfully inject video into the Ethernet.

What Network Interface Cards Do

If you’ve had any contact with computers in the past five years or so, you’ve dealt with a network card. Gone are the days, though, when you had to buy one separately from the rest of the machine. Interestingly enough, the network interface card (NIC) does just what it sounds like. It is the final interface between the computer and the physical network. If you recall Part 3 of the “D.U.M.I.E.S.” series, the word “physical” should give you an idea of what layer of the OSI model handles NIC functions. You are correct if you said Layer 1, the physical layer.

The NIC takes the data that has by now been encapsulated into Ethernet frames and converts it into electronic pulses that get sent across the wires.

Currently, the most common NICs operate at 10/100Mbps (megabits per second) and have a standard RJ-45 connector for Category-5 cable. More and more, however, we are seeing 1,000Mbps (gigabit) and fiber NICs. Be aware of the type of NIC your device has or requires.

Hubs Bring Devices Together

The word “hub” can take on a couple of meanings in a network environment. It can have a general definition of any device at the center of a star topology, or it can mean a specific device called a hub, which is how we’re going to use it.

A hub concentrates several network devices together. It’s a handy way to connect a couple of computers to share files or printers at home, but it’s not the best choice for digital video, especially in large amounts.

A hub is a very simple device. Any data that comes into any single port is repeated and sent back out through all the other ports. That’s it. It is active in the sense that some hubs regenerate the signal, but other than that, it performs a very passive function.

Unlike routers and switches, which we’ll look at momentarily, a hub has no intelligence for making switching or routing decisions. Every packet that comes in any port goes out EVERY port, period.

The biggest problem with this is the bandwidth is shared among all the ports, not dedicated to each. Someone may correctly tell you a hub is 100Mbps, but it’s for the whole device, not per port. That 100Mbps is divided among the number of ports on the device. If you were to try to run video streams through a hub, you would choke it down fairly quickly.

Could you hook one DVR up to one client computer through a hub? Sure, and it would probably work fine. But add any more clients and you’ll see trouble appear quickly.

The other big problem with hubs is security. Anyone with a network analyzer or packet-sniffing software (not hard to find) can plug into any port on a hub and see EVERY packet going through it. Remember, all packets are replicated out all ports.

You probably won’t be considering hubs for most reasonably sized video networks.

Switches Are Picky About Ports

A switch is basically a more intelligent hub, but some switches are more than that.

Originally, switches were strictly Layer 2 devices. Reviewing our last lesson, recall that Layer 2, or the Data Link Layer, gave us MAC addresses. Whenever a packet is passed down through Layer 2, a source and destination

MAC address is added through encapsulation. The resulting frame is then handed off to Layer 1. Where hubs will simply forward all data out all ports, a switch uses that MAC address to make decisions about which port the packet is destined.

When a computer is plugged into a switch, the switch records the MAC address of that computer and stores it in a MAC address table. This way, it can look at the data as it comes in and only send it to the port that needs it. This avoids flooding the network with traffic.

A big benefit to switches is since they only forward packets to the ports they were destined for (with the exception of broadcast packets), bandwidth usage is dedicated for each port. This means if a switch is designated as 100Mbits, it will have

100Mbits available for each port, not shared like the hub was.

It is also more secure in that someone connecting an analyzer or computer with sniffing software into the switch will only see data meant for that port, not everything passing through the switch (there are exceptions, but this is the general rule).

The biggest advancement in switch technology has been the introduction of multilayer switches. The most common can work at Layer 2 and Layer 3, the Network Layer. This means an L3 switch can speak the language of IP addresses and perform the same functions as a router, which we’ll discuss next.

This greatly reduces the amount of hardware necessary for a network infrastructure. L3 switches also perform those functions much faster than routers, so network latency (how long a packet takes to get from point A to B) is greatly reduced.

Virtual local area networks (VLANs) are another benefit of switches. VLANs are any number of switch ports grouped together as their own network segment. It basically takes those ports (as defined by a user) and makes them a separate network from the rest of the ports. Usually, any number of VLANs can be created on a switch. From one to all ports can be assigned to VLANs.

 VLANs can also span between multiple switches by the use of VLAN trunking protocols. Each VLAN is tagged individually so traffic from multiple VLANs can travel across a single line between switches.

A VLAN effectively reduces the number of ports that can be affected by broadcast traffic as well as the amount of broadcast traffic on the network as a whole, as that type of traffic cannot cross out of a VLAN by default. With multiple VLANs, a router is necessary to forward any traffic between VLANs, just like it would a separate physical network.

With L2 switches, an external router is necessary (networking types call this “router on a stick”). With L3 switches, the switch can perform these functions internally as it can understand IP addresses and routing protocols.

In general, high-speed switches are the future of networking and, particularly, the Internet. They are already beginning to replace the large core routers on the Internet backbone because of their speed and intelligence.

If you enjoyed this article and want to receive more valuable industry content like this, click here to sign up for our FREE digital newsletters!

Related Content