Routers are core to broadcast networks. So, I want to introduce some fundamental concepts regarding networking that may help you understand how routers work.
Ethernet is both a protocol and a hardware specification. The Ethernet specification details electrical signals and voltages on the wire (or the RF transmission scheme in the case of wireless), and it lays out how Ethernet packets are built and what information Ethernet headers contain. Headers contain source and destination addresses. This helps routers direct packets to destination. Ethernet addresses differ from IP addresses.
Ethernet addresses (sometimes called MAC addresses) appear as: nn:nn:nn:nn:nn:nn (six groups of two hex digits). These are hardcoded into network interface chips on a specific computer and typically cannot be changed. Think of MAC addresses as Vehicle Identification Numbers stamped into engine blocks of cars. Ethernet is defined by the IEEE in its 802-x family of standards.
IP is a core protocol. Its job is to move datagrams from one device to another. (A datagram is defined as a self-contained, independent entity of data. It carries sufficient information to be routed from a source to the destination computer without reliance on earlier exchanges between the source and destination computer and transporting network.) The IP layer prepares data sent to it by applications or other higher protocols for transmission across a specific network (an IP network), taking into account things such as packet length, hardware addressing structure and how data should split across multiple packets.
IP packets are the payload carried in Ethernet packets in an “IP over Ethernet” network (about 95 percent of all network installations). MAC addresses may be hardcoded, but IP addresses can be changed in the network configuration menu of almost all modern operating systems. IP is defined by the Internet Engineering Task Force (IETF) as Internet Standard 5.
Once a physical network is in place (wires, connectors, switches, etc.), equipment can be connected to it. With Ethernet and IP, the capability exists to identify specific equipment as packet sources and destinations. And, the means are available to logically group equipment into networks and pass messages from one network to another. But, another protocol is critical to the network functioning.
Here is a question: What mechanism associates the MAC address and, therefore, a specific physical piece of hardware, with an IP address?
Address Resolution Protocol, or ARP, provides the solution. Figure 1 shows an actual ARP transaction captured using Wireshark — a free packet capture and inspection tool. Assume a router with an IP address of 192.168.1.1 needs to send a packet to 192.168.1.43. The router will send an ARP request asking, “Who has 192.168.1.43? Tell 19.2168.1.1.” The computer with that IP address responds, “192.168.1.43 at 00:15:53:7C:22:5C.” The router then knows where the packet should go. In most cases, once the router has this information, it stores it in its ARP table for memory. ARP is defined in IETF Internet Standard 37.
Figure 1. A Wireshark capture shows ARP at work resolving a MAC address to an IP address.
Now that the IP address to MAC address question has been resolved, the network is almost ready to go. Packets can be sent from one computer to another on the same network, but what about from one network to another?
Now, we will talk about routes and route tables and how they relate to IP networks. To do this, we need to understand what a network is in terms of IP. For example, let’s say you are responsible for designing a network in your station. This network will connect to the news and production departments.
Both news and production have their own servers and printers, and you do not want traffic from the production department slowing down news and vice versa.
This is a perfect situation for a network. You assign production computer addresses in the form 192.168.0.n, and news 192.168.1.n. Now, you have two networks on one wire. But, if these are two networks, and computers ignore traffic not on the assigned network, how do they communicate?
A router needs to send packets from one network to another. The router keeps a route table that tells it where to send different types of traffic.
Every IP device has one. In this simple example, if on a news computer you open a terminal window and type “route print” (or a similar command, depending on the system), you will get output like that shown in Table 1.
Table 1. A routing table charts commands for a network router to follow. Here, this table shows the connection between news and production department networks.
This table says if this news computer has traffic for 192.168.1.n (the news network), then keep it local and do not send it elsewhere. This is shown by the asterisk (*) in the Gateway entry for 192.168.1. But, if traffic exists for any other network, send it to the router at 192.168.0.1, and that router will forward traffic where it needs to go. The router also contains a route table, which will be more complicated than this one, that tells it what to do with various packets depending on the destination address. Understanding route tables is important in troubleshooting networks.
So far, everything we have covered is centered on packets and addresses. But, humans are using computer systems, and humans do not like long numbers. We would much rather use memorable names. For example, if we are looking for Broadcast Engineering’s website, it is easy for us to remember http://www.broadcastengineering.com rather than http://220.127.116.11. The Domain Name System, or DNS, resolves names into IP address.
In a nutshell, DNS is a distributed database. When you enter http://www.broadcastengineering.com, the browser asks a DNS server what IP address to go to. The server replies with 18.104.22.168.
This is important because DNS problems are reported frequently as network outages. Before looking for network problems, try checking the network by typing an IP address into a browser rather than a name. If the page loads, you may have a DNS server issue rather than a connectivity concern. They can look the same.
User Datagram Protocol (UDP) is a straight-forward protocol that allows for mapping of video onto IP. UDP is included here because routers frequently are configured to block UDP traffic. This is because hackers can use it to crash networks.
There are good reasons for UDP in professional media applications, however. So, each user must balance security policies and user requirements. With a UDP problem, be sure network routers and switches are not blocking UDP.
—Brad Gilmer is President of Gilmer & Associates, executive director of the Advance Media Workflow Association and executive director of the Video Services Forum.