1 / 10

IPPM Delay Variation Applicability Statement

This draft discusses the applicability and uses of the Delay Variation metric in network characterization and application performance estimation. It also covers measurement considerations and provides guidance for reducing delay variation in networks.

josephyoung
Télécharger la présentation

IPPM Delay Variation Applicability Statement

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Delay Variation Applicability Statementdraft-morton-ippm-delay-var-as-03 July 24, 2007 Al Morton Benoit Claise

  2. Applicability Problem: “How will the results be used?” • Krzanowski introduced the Delay Variation Problem at IETF-64 • “How” Question asked at IETF-65, no suggestions yet • RFC 3393 lists two key uses for the Delay Variation Metric • Memo Considers these 3 Tasks and N Special Circumstances Network Characterization: Inferring Queue Occupation on a Path Application Performance Estimation: Sizing of De-Jitter Buffers What about Spatial Composition ? DST SRC What if circumstances are challenging?

  3. Inter-Packet Delay Var. (selection func = previous packet) Src Dst T1 IPDV(2) = (R2-R1) – (T2-T1) R1 T2 R2 T3 R3 T4 (5)(2) (4) (3) T5 R4 IPDV(4) = (R4-R3) – (T4-T3) R5

  4. Packet Delay Variation(selection func = minimum delay pkt in stream) Src Dst T1 PDV(3) = (R3-T3) – (R2-T2) R1 T2 (1,3,5) R2 (minimum) T3 R3 T4 (2) (4) T5 R4 PDV(4) = (R4-T4) – (R2-T2) R5

  5. 1.  Introduction1.1.  Background Literature in IPPM and Elsewhere1.2.  Organization of the Memo 2.Purpose and Scope 3.Brief Descriptions of Delay Variation Uses (four) 4.Formulations of IPDV & PDV 5.Earlier Comparisons 6.Additional Properties and Comparisons6.1.  Packet Loss6.2.  Path Changes6.2.1.  Lossless Path Change6.2.2.  Path Change with Loss6.3.  Clock Stability and Error6.4.  Spatial Composition6.5.  Reporting a Single Number6.6.  Jitter in RTCP Reports6.7.  MAPDV26.8.  Load Balancing 7.Applicability of the Delay Variation Forms and Rec7.1.  Uses7.2.  Challenging Circumstances7.2.1.  Clock Issues7.2.2.  Frequent Path Changes7.2.3.  Frequent Loss7.2.4.  Load Balancing 8.Measurement Considerations for Vendors, Testers, and Users8.1.  Measurement Stream Characteristics8.2.  Measurement Units8.3.  Test Duration8.4.  Clock Sync Options8.5.  Distinguishing Long Delay from Loss8.6.  Accounting for Packet Reordering8.7.  Results Representation and Reporting… 12.Appendix on Reducing Delay Variation in Networks 13.Appendix on Calculating the D(min) in PDV Current Outline of the Draft

  6. Recap of activity: IETF-68 and Updates in 03 • Strong support for this work and its current conclusions during the IPPM Session • No opposition, either • Key “scope/coverage” questions raised, but no feedback • Completed Sec 4 on the initial comparison between IPDV and PDV, including example, graphs and table • Completed Sec 13 Appendix on Calculating the minimum delay reference, D(min) for PDV • Looked at the “Video Packet Smoothing Buffer” problem with IPDV and PDV – no conclusion yet

  7. Summary of Comparisons • Challenging Circumstances for measurement: • IPDV form offers advantages when • Path changes are Frequent • Meas. System Clocks exhibit significant Skew • PDV form is less sensitive to Packet Loss • Spatial Composition of DV metric: • All known methods use PDV, • IPDV sensitivity to sequence is an issue • Estimate of Queuing Time & Variation: • PDV estimates this, especially when sample min = true min • Determine De-jitter Buffer Size Required • PDV “pseudo-range” reveals this property by anchoring the distribution at the minimum delay

  8. What if there is load-balancing in the network? Load- balancing is based on the IGP metrics, while the delay depends on the path. So, we could have a multi-modal distribution, if we send test packets with different characteristics such as IP addresses/ports. • Should the delay singletons using multiple addresses/ports be combined in the same sample? The PDV form makes the identification of multiple modes possible. • Should we characterize each mode separately? • This question also applies to the Path Change case.

  9. New Sections (02) and Questions!!! • Section 8: Measurement Considerations for Vendors, Testers, and Users • This section provides guidance in many areas – Useful? • Do we want to recommend measuring BOTH IPDV and PDV? • Do we want to reconstruct the IPDV and DV later on, with a different interval? • Appendix: Guidance for reducing DV in networks • Useful?

  10. Summary • IPDV and PDV each have their Strengths and Weaknesses • Initial (00) Conclusions have seen agreement • Suggestions for additional Tasks & Circumstances • Need Feedback on Questions & proposed sections and questions raised in the text of the draft • Need readers and comments to help finish this up • Charter Item Time…

More Related