1. A Bandwidth Estimation Method for IP Version 6 Networks Marshall Crocker Department of Electrical and Computer Engineering Mississippi State University October 13, 2006

2. Page 2 of 37 MS Thesis Presentation Outline Introduction to Bandwidth and Estimation Motivation IPv4 Estimation Techniques IPv6 Overview and Estimation Technique Simulation Experimentation Conclusions and Future Work

3. Page 3 of 37 MS Thesis Presentation What is Bandwidth? Important characteristic of data networks  How much data  How fast Determined by two primary properties  Physical Link Capacity  Infrastructure Utilization Data? 100 Mb/s 10 Mb/s % Load

4. Page 4 of 37 MS Thesis Presentation What is Bandwidth Estimation? Nodes attempt to estimate network bandwidth  Determine minimum physical capacity called Bottleneck Bandwidth  Determine unused capacity called Available Bandwidth Estimations are used in many different ways Many different techniques for performing estimations 100 Mb/s 10 Mb/s 90% Load

5. Page 5 of 37 MS Thesis Presentation Sending rate less than BWE? Applications of BWE End-to-end flow control Increase Rate Decrease Rate yes no

6. Page 6 of 37 MS Thesis Presentation Applications of BWE Server selection for downloads and streaming media Peer-to-peer selection  Connect to peers with most bandwidth Traffic Engineering  Adjust routing/switching for optimal operation Capacity Provisioning  Increase/decrease capacity as needed 10 5 8

7. Page 7 of 37 MS Thesis Presentation Motivation BWE valuable for a number of applications IPv4 techniques  Suffer from various flaws  Limited due to nature of the network  No single technique suitable for all applications  Evolving network technologies affect theories IPv6 offers framework for improved estimation technique  Efficient  Flexible  Accurate  Simple

8. Page 8 of 37 MS Thesis Presentation IPv4 Estimation Techniques All measurements are passive Examine how network delivers data  Packet spacing  Packet delay  Packet dispersion  Statistical

9. Page 9 of 37 MS Thesis Presentation IPv4 Estimation Techniques Each suffer from one or more  High level of complexity  Poor efficiency  Limited accuracy  Application specificity Each method is susceptible to one or more  Network load  Cross-traffic  Packet-size variability  Probing packet size  Train length  Cross-traffic routing

10. Page 10 of 37 MS Thesis Presentation Incorrect IPv4 Bandwidth Estimation Scenario

11. Page 11 of 37 MS Thesis Presentation IP Version 6 Next generation Internet Protocol Improves on IPv4  Expanded addressing from 32 bits to 128  Simplified header  Improved extension and option support Extension support provides framework for improved bwe technique

12. Page 12 of 37 MS Thesis Presentation Data …. IPv6 Header Ver 6Traffic ClassFlow Label 32 bits Payload LengthNext Hdr.Hop Limit Source Address Destination Address Extensions

13. Page 13 of 37 MS Thesis Presentation IPv6 Extensions Several different extensions  Routing  Fragmentation  Destination options  Authentication  Security  Hop-by-hop Examined by every hop Provides instructions for each hop Only two options currently defined  Jumbo payload  Router alert

14. Page 14 of 37 MS Thesis Presentation Proposed Hop-by-Hop Options Traceroute  Each hop inserts address  Record forward/backward path  Not accepted by IETF Connection Status Investigation (CSI)  Request statistics/attributes for each hop  IP address  Bandwidth  Type  Number of transmitted/received bytes/packets  Number of errors

15. Page 15 of 37 MS Thesis Presentation IPv6 Timestamp Option CSI would have been extremely useful Rejected by IETF due to complexity, security, and proprietary concerns A timestamp option was defined for IPv4 but had limited use An IPv6 timestamp option has much more potential including bandwidth estimation

16. Page 16 of 37 MS Thesis Presentation IPv4 Timestamp Deficiencies IPv4 timestamp option limited in usefulness  Can only hold timestamps for up to 9 hops without addresses  Room to hold 4 hops with addresses  No standard for defining timestamp format  IPv4 routers services packets with options slower

17. Page 17 of 37 MS Thesis Presentation IPv6 Timestamp IPv6 timestamp properties  Enough room to hold timestamp records for every hop  Predefined timestamp format  Timestamp at incoming and/or outgoing interfaces

18. Page 18 of 37 MS Thesis Presentation IPv6 Timestamp Format Next Hdr.Hdr. Ext LenOption TypeOption Data Len 32 bits Record CountTS TypeHop Limit BaseResIfOptR IdentifierReserved Upper Part of IPv6 Address Lower Part of IPv6 Address Hop NumberCounterLk TypeI/FResolutionG TimestampFmt

19. Page 19 of 37 MS Thesis Presentation IPv6 Timestamp BWE Define bandwidth as number of transmitted bits per unit time Expand to include start and end transmit times Use start/end transmit times of packet and packet size to calculate capacity Send two timestamp packets back-to-back Timestamp of first packet and timestamp of second packet = t1 and t2 Size of first packet and link layer size used in final calculation

20. Page 20 of 37 MS Thesis Presentation Bottleneck Bandwidth Estimation LeadTail Router TS = 10TS = 15

21. Page 21 of 37 MS Thesis Presentation Bottleneck Bandwidth Estimation Relies on back-to-back queuing Count field in TS record ensures back-to-back Smaller tail packet helps back-to-back queuing

22. Page 22 of 37 MS Thesis Presentation Available Bandwidth Estimation LeadTail Router TS = 10TS = 20 CT

23. Page 23 of 37 MS Thesis Presentation Available Bandwidth Estimation Relies on cross traffic to introduce packet separation Constantly changing value Applications must send estimations frequently

24. Page 24 of 37 MS Thesis Presentation Simulation Experimentation Simulation experiments used to compare and evaluate IPv6 Timestamp method Measured against comparable IPv4 method called the cartouche method Cartouche method uses packet trains and examines packet spacing to estimate BW

25. Page 25 of 37 MS Thesis Presentation Simulation Setup

26. Page 26 of 37 MS Thesis Presentation Simulation Setup ParameterValues Number of Cross Traffic Flows8, 16, 24, 32, 40, 48 Cross Traffic Rates100 Kbps, 1 Mbps Average Cross Traffic Size200 Bytes Leading Probe Packet Size1500 Bytes Bottleneck Rates (Mbps)10, 20, 30, 40, 50, 60, 70, 80, 90, 100 Hop Rates100 Mbps

27. Page 27 of 37 MS Thesis Presentation Estimation Method Parameters ScenarioCT RateTailgating Packet Size (Bytes) 1100 Kbps212 2100 Kbps76 31 Mbps212 41 Mbps76 ScenarioCT RateTailgating Packet Size (Bytes) Cartouche Length (r) 1100 Kbps402 2100 Kbps403 31 Mbps402 41 Mbps403

28. Page 28 of 37 MS Thesis Presentation IPv6 Estimation Results Scenario 1 Scenario 4

29. Page 29 of 37 MS Thesis Presentation Cartouche Estimation Results Scenario 1 Scenario 4

30. Page 30 of 37 MS Thesis Presentation IPv6 Estimation Frequency

31. Page 31 of 37 MS Thesis Presentation IPv6 Estimation Frequency

32. Page 32 of 37 MS Thesis Presentation Cartouche Estimation Frequency

33. Page 33 of 37 MS Thesis Presentation Cartouche Estimation Frequency

34. Page 34 of 37 MS Thesis Presentation Cartouche Estimation Frequency

35. Page 35 of 37 MS Thesis Presentation Cartouche Estimation Frequency

36. Page 36 of 37 MS Thesis Presentation Conclusions Presented IPv6 bandwidth estimation using timestamp hop- by-hop option Advantageous over existing methods  Efficient  Simple  Flexible  Accurate IPv4 bandwidth estimations are limited due to the nature of the network Outperforms comparable IPv4 Technique

37. Page 37 of 37 MS Thesis Presentation Future Work Extended simulation models  Diverse network properties and conditions  Additional hardware and communications models  Additional host and network models Real world implementation Development of network control techniques, protocols and applications such as a “Cognizant” version of TCP  Aware of network  Intelligently respond to network and conditions  Fairly use network resources