Traffic Model and Traffic Model and Engineering Engineering Dr. - PowerPoint PPT Presentation

Acknowledgement/Reference Acknowledgement/Reference � Slides are taken from the following source: Slides are taken from the following source: � CS 5224 CS 5224 � S. S. Keshav Keshav, “An Engineering Approach to Computer , “An Engineering Approach to Computer � High Speed Networks and Multimedia High Speed Networks and Multimedia Networking”, Chapter 14: Traffic Management Networking”, Chapter 14: Traffic Management Networking Networking Traffic Model and Traffic Model and Engineering Engineering Dr. Chan Mun Choon Dr. Chan Mun Choon School of Computing, National University of Singapore School of Computing, National University of Singapore August 17, 2005 (week 2/3) Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 2 1 Motivation for Traffic Models An example Motivation for Traffic Models An example � In order to predict the performance of a In order to predict the performance of a � Executive participating in a worldwide � network system, we need to be able to network system, we need to be able to videoconference “describe” the “behavior” of the input traffic “describe” the “behavior” of the input traffic � Proceedings are videotaped and stored in an � Often, in order to reduce the complexity, we classify Often, in order to reduce the complexity, we classify � archive the user behavior into classes, depending on the the user behavior into classes, depending on the applications applications � Edited and placed on a Web site � Sometimes, we may be even able to “restrict” or Sometimes, we may be even able to “restrict” or � � Accessed later by others shape the users’ behavior so that they conform to shape the users’ behavior so that they conform to � During conference some specifications some specifications � Only when there is a traffic model is traffic Only when there is a traffic model is traffic � Sends email to an assistant � engineering possible engineering possible � Breaks off to answer a voice call Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 3 Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 4 1

What this requires Traffic management What this requires Traffic management � For video � Set of policies and mechanisms that allow a � sustained bandwidth of at least 64 kbps network to efficiently satisfy a diverse range of � low loss rate service requests � For voice � Tension is between diversity and efficiency � sustained bandwidth of at least 8 kbps � Traffic management is necessary for providing � low loss rate � For interactive communication Quality of Service (QoS) � low delay (< 100 ms one-way) � Subsumes congestion control (congestion == loss of � For playback efficiency) � low delay jitter � For email and archiving � reliable bulk transport Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 5 Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 6 Time Scale of Traffic Management Time Scale (cont.) Time Scale of Traffic Management Time Scale (cont.) � Session (call-level) � Less than one round-trip-time (cell-level) � End-points interact with network elements � Perform by the end Perform by the end- -points and switching nodes points and switching nodes � � Signaling Signaling � � Scheduling and buffer management Scheduling and buffer management � � Admission control Admission control � � Regulation and policing Regulation and policing � Service pricing Service pricing � � � Routing (connection Routing (connection- -oriented networks) oriented networks) � Policy routing (datagram networks) Policy routing (datagram networks) � � � Day � One or more round-trip-times (burst-level) � Human intervention � Perform by the end-points � Peak load pricing Peak load pricing � � Weeks or months � Feedback flow control Feedback flow control � � Human intervention � Retransmission Retransmission � � Capacity planning Capacity planning � � Renegotiation Renegotiation � Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 7 Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 8 2

Some economic principles Principles applied Some economic principles Principles applied � A single network that provides heterogeneous QoS is A single network that provides heterogeneous QoS is � A single wire that carries both voice and data is more A single wire that carries both voice and data is more � � better than separate networks for each QoS efficient than separate wires for voice and data better than separate networks for each QoS efficient than separate wires for voice and data � unused capacity is available to others unused capacity is available to others � ADSL ADSL � � � Lowering delay of delay Lowering delay of delay- -sensitive traffic increased sensitive traffic increased � IP Phone IP Phone � � welfare welfare � Moving from a 20% loaded10 Mbps Ethernet to a 20% Moving from a 20% loaded10 Mbps Ethernet to a 20% � loaded 100 Mbps Ethernet will still improve social loaded 100 Mbps Ethernet will still improve social � can increase welfare by matching service menu to user can increase welfare by matching service menu to user � requirements requirements welfare welfare � BUT need to know what users want (signaling) BUT need to know what users want (signaling) � � increase capacity whenever possible increase capacity whenever possible � � Better to give 5% of the traffic lower delay than all Better to give 5% of the traffic lower delay than all � traffic low delay traffic low delay � should somehow mark and isolate low should somehow mark and isolate low- -delay traffic delay traffic � Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 9 Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 10 The two camps Telephone traffic models (Call) The two camps Telephone traffic models (Call) � How are calls placed? How are calls placed? � Can increase welfare either by Can increase welfare either by � � � call arrival model call arrival model � matching services to user requirements matching services to user requirements or or � � � studies show that time between calls is drawn from an studies show that time between calls is drawn from an � increasing capacity blindly increasing capacity blindly � � exponential exponential distribution distribution � Which is cheaper? Which is cheaper? � � call arrival process is therefore call arrival process is therefore Poisson Poisson � � depends on technology advancement depends on technology advancement � � memoryless memoryless: the fact that a certain amount of time has passed : the fact that a certain amount of time has passed � � User behavior/expectation/tolerance User behavior/expectation/tolerance � since the last call gives no information of time to next call since the last call gives no information of time to next call � small and smart vs. big and dumb small and smart vs. big and dumb � � How long are calls held? How long are calls held? � � It seems that smarter ought to be better It seems that smarter ought to be better � � usually modeled as usually modeled as exponential exponential � � otherwise, to get low delays for some traffic, we need to give otherwise, to get low delays for some traffic, we need to give all traffic all traffic low low � � however, measurement studies (in the mid however, measurement studies (in the mid- -90s) show that it is 90s) show that it is � delay, even if it doesn’t need it delay, even if it doesn’t need it heavy tailed heavy tailed � But, perhaps, we can use the money spent on traffic But, perhaps, we can use the money spent on traffic � A small number of calls last a very long time A small number of calls last a very long time � � � Why? Why? management to increase capacity management to increase capacity � � We will study traffic management, assuming that it matters! We will study traffic management, assuming that it matters! � Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 11 Aug 17, 2005 (Week 2/3) Traffic Model/Engineering 12 3