Presentation is loading. Please wait.

Presentation is loading. Please wait.

JTR04, Montpellier, France, 4-6 octobre 2004 1 Traffic characterization and analysis Nicolas Larrieu, Philippe Owezarski LAAS-CNRS Toulouse, France {nlarrieu,

Similar presentations


Presentation on theme: "JTR04, Montpellier, France, 4-6 octobre 2004 1 Traffic characterization and analysis Nicolas Larrieu, Philippe Owezarski LAAS-CNRS Toulouse, France {nlarrieu,"— Presentation transcript:

1 JTR04, Montpellier, France, 4-6 octobre 2004 1 Traffic characterization and analysis Nicolas Larrieu, Philippe Owezarski LAAS-CNRS Toulouse, France {nlarrieu, owe}@laas.fr

2 JTR04, Montpellier, France, 4-6 octobre 2004 2 Outline Introduction to Internet monitoring Needs, principles and applications State of the art on active and passive technics Internet traffic characterisitics and related performance problems Traffic mutation: P2P impact Treats for the Internet: oscillations, LRD… How to use monitoring results? Application to traffic engineering: a new congestion control mechanism

3 JTR04, Montpellier, France, 4-6 octobre 2004 3 Part 1 Introduction to Internet monitoring Needs, principles and applications State of the art on active and passive technics Internet traffic characterisitics and related performance problems Traffic mutation: P2P impact Treats for the Internet: oscillations, LRD… How to use monitoring results? Application to traffic engineering: a new congestion control mechanism

4 JTR04, Montpellier, France, 4-6 octobre 2004 4 Introduction Deals with both monitoring results and effects on network design, research and management Topic under the spotlight Definition: « science of measurements » To apply to networks in general, to the Internet in particular

5 JTR04, Montpellier, France, 4-6 octobre 2004 5 Monitoring and networks… Explosion of the Internet (in size) New services QoS … Increasing complexity of the Internet No more control of the global Internet

6 JTR04, Montpellier, France, 4-6 octobre 2004 6 Monitoring and networks… « Defeat » statement For instance on QoS aspects Bad knowledge of the traffic Because of the current Internet complexity, its evolution requires a good knowledge of the « real » traffic « Real » traffic exists and is full of information

7 JTR04, Montpellier, France, 4-6 octobre 2004 7 Example of a provisioning problem

8 JTR04, Montpellier, France, 4-6 octobre 2004 8 What to use for network monitoring? Administration / operation tools based on SNMP Topology of networks / configuration Some statistics measurements Granularity is too coarse: min = 5 s (but can be 1 hour, 1 day, 1 week or whatever) Measured parameters are more or less the amount of traffic sent and received

9 JTR04, Montpellier, France, 4-6 octobre 2004 9 Some examples of SNMP results (1) Per hour trace Input traffic Output traffic RAP RENATER interconnection

10 JTR04, Montpellier, France, 4-6 octobre 2004 10 Some examples of SNMP results (2) Per Week trace Per Month trace Input traffic Output traffic

11 JTR04, Montpellier, France, 4-6 octobre 2004 11 Problems for monitoring networks Impossible to monitor traffic dynamics (second order values as variability auto- covariance for instance) Impossible to monitor traffic QoS (user point of view – goodput) Impossible to get a (formal) traffic model

12 JTR04, Montpellier, France, 4-6 octobre 2004 12 Example on network provisioning Common beliefs tell us traffic is Poisson: E[X]= V[X]= Provisioning should be 2 Actually, provisioning has to be at least 1:3 (i.e. 3 ) RENATER 1:3 Sprint 1:3 WorldCom 1:5 AT&T 1:10

13 JTR04, Montpellier, France, 4-6 octobre 2004 13 Questions on the example How explaining this over-provisioning requirement ? How to predict the traffic that will be supported by a new network to design ? ?

14 JTR04, Montpellier, France, 4-6 octobre 2004 14 Part 2 Introduction to Internet monitoring Needs, principles and applications State of the art on active and passive technics Internet traffic characterisitics and related performance problems Traffic mutation: P2P impact Treats for the Internet: oscillations, LRD… How to use monitoring results? Application to traffic engineering: a new congestion control mechanism

15 JTR04, Montpellier, France, 4-6 octobre 2004 15 Active measurements Principles Consists in sending packets on a network and observing results (Delay, RTT, Throughput, etc.) User point of view Best solution to evaluate the service you can get from the network youre connected to Drawbacks Probe packets change the state of the network IETF IPPM WG is working on the definition of probing scenarios minimizing the effects on the network state

16 JTR04, Montpellier, France, 4-6 octobre 2004 16 Some active measurement tools Ping Traceroute MGEN RIPE equipments Etc. Importance of clock synchronization: most of the time GPS is required

17 JTR04, Montpellier, France, 4-6 octobre 2004 17 Passive measurements Principles Capture packets (or headers) Not intrusive at all Carrier / ISP point of view Best solution for a carrier to measure traffic Drawbacks Sampling issues Difficult to get a user point of view Technical limits (speed of components, capacity)

18 JTR04, Montpellier, France, 4-6 octobre 2004 18 On line vs. Off line measurements On line Packets are analyzed in real-time Analysis on very long periods But complexity of analysis is quite limited Off line Packets are stored on hard drives / SAN for later analysis Possibilities of analysis are endless Possibility of correlating several traces But amount of stored data is really huge (small periods only)

19 JTR04, Montpellier, France, 4-6 octobre 2004 19 Some passive measurement tools TSTAT LIBCAP Tcpdump Tcptrace QoSMOS IPANEMA CISCOs Netflow DAG cards Etc.

20 JTR04, Montpellier, France, 4-6 octobre 2004 20 Part 3 Introduction to Internet monitoring Needs, principles and applications State of the art on active and passive technics Internet traffic characterisitics and related performance problems Traffic mutation: P2P impact Treats for the Internet: oscillations, LRD… How to use monitoring results? Application to traffic engineering: a new congestion control mechanism

21 JTR04, Montpellier, France, 4-6 octobre 2004 21 Current Internet traffic New applications with various and changing requirements in terms of QoS appear New P2P applications make traffic characteristics change

22 JTR04, Montpellier, France, 4-6 octobre 2004 22 Internet traffic evolution (May 2000)

23 JTR04, Montpellier, France, 4-6 octobre 2004 23 Internet traffic evolution (August 2000)

24 JTR04, Montpellier, France, 4-6 octobre 2004 24 Internet traffic evolution (May 2003) Main TCP applications throughputs (Renater) Time Throughput (kbits/s)

25 JTR04, Montpellier, France, 4-6 octobre 2004 25 TCP flow size Number of packets per flow Percentage 0 10 20 30 40 50 60 70 80 90 0-1010-100 >100

26 JTR04, Montpellier, France, 4-6 octobre 2004 26 TCP flow size vs. total bandwidth Number of packets per flow Percentage 0 10 20 30 40 50 60 70 80 90 0-1010-100 >100 flows bytes « Mice vs. elephants »

27 JTR04, Montpellier, France, 4-6 octobre 2004 27 Impact of P2P on traffic Thousands of mice A large number of elephants Change flow size distribution

28 JTR04, Montpellier, France, 4-6 octobre 2004 28 Flow size distribution Exponential August 2000 May 2003 % Nb of packets / flow 40201008060 20 40 60 80 100 90 70 50 30 10

29 JTR04, Montpellier, France, 4-6 octobre 2004 29 Traffic oscillation issues

30 JTR04, Montpellier, France, 4-6 octobre 2004 30 Illustration: LRD and losses

31 JTR04, Montpellier, France, 4-6 octobre 2004 31 Illustration: LRD and losses

32 JTR04, Montpellier, France, 4-6 octobre 2004 32 Illustration: LRD and losses

33 JTR04, Montpellier, France, 4-6 octobre 2004 33 Illustration: LRD and losses

34 JTR04, Montpellier, France, 4-6 octobre 2004 34 Illustration: LRD and losses

35 JTR04, Montpellier, France, 4-6 octobre 2004 35 Wavelett analysis of the traffic Range Time

36 JTR04, Montpellier, France, 4-6 octobre 2004 36 Log-log scale variance diagram 2 different power laws 2 different rating for scale invariance High scales H 0.5 LRD characteristics

37 JTR04, Montpellier, France, 4-6 octobre 2004 37 Oscillations persistence characterization H = 0.741

38 JTR04, Montpellier, France, 4-6 octobre 2004 38 Traffic characteristics bibliography Actual Internet traffic has self-similarity and long range dependancy properties Due to: Heavy-tailed distribution of flow size TCP-like congestion control mechanism Routers Human and application behaviors

39 JTR04, Montpellier, France, 4-6 octobre 2004 39 Self-similarity

40 JTR04, Montpellier, France, 4-6 octobre 2004 40 Illustration of self-similarity Worst case with a highly self-similar traffic E[X] = V[X] O( 2 ) Example: If E[X] = 100 Mbps Provisioning for a reliable network with low delays is in the order of 10 Gbps Reach the technological limits of networks with an average traffic of few hundreds of Mbps

41 JTR04, Montpellier, France, 4-6 octobre 2004 41 Conclusion on traffic evolution Actual Internet traffic is not Poisson Well, it is probably not self-similar at all scales? But it is really LRD (and this is a major issue!)

42 JTR04, Montpellier, France, 4-6 octobre 2004 42 Part 4 Introduction to Internet monitoring Needs, principles and applications State of the art on active and passive technics Internet traffic characterisitics and related performance problems Traffic mutation: P2P impact Treats for the Internet: oscillations, LRD… How to use monitoring results? Application to traffic engineering: a new congestion control mechanism

43 JTR04, Montpellier, France, 4-6 octobre 2004 43 Measurement Based… Monitoring/measurements become the corner stone of research in networking Many topics are addressed: New protocols and architectures for: Traffic characterization and modeling Multi-domains QoS guaranty Service and network utilization optimization Network or VPN or CoS provisioning QoS routing Network security, etc. Techniques and mechanisms for: Pricing, etc.

44 JTR04, Montpellier, France, 4-6 octobre 2004 44 Multiple causes for Internet oscillations 1. TCP like congestion control mechanisms (Slow Start and Congestion Avoidance mechanisms / Closed control loop) 2. Increase of transmitted file size 3. Increase of network capacities (and over- provisioning) Increase of oscillations Amplitude Range traffic oscillations limit network performance « High variability » paper of Willinger (IEEE ToN 96)

45 JTR04, Montpellier, France, 4-6 octobre 2004 45 Smoothing flow behavior Disturbances are mainly due to elephants We need to increase elephant flows regularity We are going to use the TFRC mechanism to transmit elephant flows TFRC proposes a smooth sending rate with very soft increases and decreases

46 JTR04, Montpellier, France, 4-6 octobre 2004 46 TFRC impact on flow QoS: throughput analysis ProtocolAverage troughput (kB)Throughput σ (kB) TCP New Reno (NR): real case TCP NR & TFRC: simulated case 82.335 77.707 157.959 102.176

47 JTR04, Montpellier, France, 4-6 octobre 2004 47 TFRC impact on flow QoS: LRD analysis Real traffic Simulated traffic

48 JTR04, Montpellier, France, 4-6 octobre 2004 48 Conclusion Monitoring give us very important information on traffic characteristics and related performance problems Traffic engineering can be really improve with monitoring results Lots of network research fields can be investigated and improved QoS, security, pricing, networking…

49 JTR04, Montpellier, France, 4-6 octobre 2004 49 Thank you for your attention Any questions? Contact: nlarrieu@laas.fr


Download ppt "JTR04, Montpellier, France, 4-6 octobre 2004 1 Traffic characterization and analysis Nicolas Larrieu, Philippe Owezarski LAAS-CNRS Toulouse, France {nlarrieu,"

Similar presentations


Ads by Google