Latency versus Bandwidth... What is it?

One of the most commonly misunderstood concepts in networking is speed and capacity. Most people believe that capacity and speed are the same thing.  For example, it's common to hear "How fast is your connection", invariably, the answer will be "640K, 1.5M" or something similar. These answers are actually referring to the bandwidth or capacity of the service, not speed.

Speed (latency) and capacity (bandwidth) are two very separate things. The combination of latency and bandwidth give users the perception of how quickly a webpage loads or a file is transferred.  It doesn't help that broadband providers keep referring to "get high speed access" when they probably should be saying "get high capacity access". Notice the term "Broadband", it refers to how wide the pipe is, not how fast.

The most common example to compare latency and bandwidth is: Imagine water running through a pipe. The pressure is latency, the width of the pipe is bandwidth. If you have a wide pipe but low pressure, you can move more water through the pipe but at a slower rate. If you have a narrow pipe but high pressure, you can move less water but at a faster rate.

Another example that is sometimes given: Imagine people in an aircraft. In this example, people are the data packets, the size of the aircraft is the bandwidth, and the speed of the aircraft is the latency.  A 747 can carry about 400 people but a 707 can carry only 200 people. Both fly at about 500 knots. If both leave New York at the same time, they will arrive in Los Angeles at the same time. Notice that although, the 747 has more capacity (or bandwidth) it is the same speed (latency) as the 707.

Latency is normally expressed in milliseconds. One of the most common methods to measure it is use the utility ping.  A small packet of data, typically 32 bytes, is sent to a host and the time is measured. Normally, the RTT (round-trip time it takes for the packet to leave the source host, travel to the destination host and return back to the source host) is measured.

Bandwidth is normally expressed in bits per second.  It's the amount of data that can be transferred during a second.  Solving bandwidth is easier than solving latency. To solve bandwidth, more pipes are added. For example, in early analog modems it was possible to increase bandwidth by bonding two or more modems. In fact, ISDN achieves 128K of bandwidth by bonding two 64K channels using a data link protocol called multilink-ppp.

The following are typical latencies, to the first hop, by popular circuit types . Please remember however that latency on the Internet is also effected by routing (number of hops) that an ISP may perform (i.e., if your data packet has to travel farther, latencies increase).

Ethernet:  0.3ms
Analog Modem:  100-200ms
ISDN:  15-30ms
DSL/Cable:  10-20ms
Satellite:  >100ms
DS1/T1    ;  2-5ms

Bandwidth and latency are connected. If the bandwidth is saturated then congestion occurs and latency is increased. However, if the bandwidth of a circuit is not at peak, the latency will not decrease. Bandwidth can always be increased but latency cannot be decreased. Latency is the function of the electrical characteristics of the circuit. So, no matter how un-congested the analog modem line is, the latency (speed) will not be reduced (increased).