Chapter 6. Traffic Policing and Shaping
QoS Exam Objectives
This chapter covers the following exam topics specific to the QoS exam:
Describe the purpose of traffic conditioning using traffic policing and traffic shaping and differentiate between the features of each Explain how network devices measure traffic rates using single rate or dual rate, single or dual token bucket mathematical models Identify the Cisco IOS commands required to configure and monitor single rate and dual rate CB-Policing Identify the Cisco IOS commands required to configure and monitor percentage based CB-Policing Explain how the two rate limits, average rate and peak rate, can be used to rate limit traffic Identify the Cisco IOS commands required to configure and monitor CB-Shaping Identify the Cisco IOS commands required to configure and monitor Frame Relay adaptive CB-Shaping on Frame Relay interfaces
Traffic policing allows devices in one network to enforce a traffic contract. Traffic contracts define how much data one network can send into another, typically expressed as a committed information rate (CIR) and a committed burst (Bc). Policing measures the flow of data, and discards packets that exceed the traffic contract.
Similarly, traffic shaping allows packets to conform to a traffic contract. In cases where packets that exceed the traffic contract might be discarded, the sending device may choose just to slow down its sending rate, so that the packets are not discarded. The process of sending the traffic more slowly than it could be sent, to conform to a traffic contract, is called shaping.
In short, policing typically drops out-of-contract traffic, whereas shaping typically delays it.
Shaping and policing share several concepts and mechanisms. Both need to measure the rate at which data is sent or received, and take action when the rate exceeds the contract. Often when policing is used for packets entering a network, shaping is also used on devices sending into that network. Although shaping and policing are not always used in the same networks, there are more similarities than differences, so both are covered in this single chapter.
 |