Abstract

Multicast applications and network monitors can potentially benefit from the ability to infer the loss rates along links within a multicast tree. Estimators, known generically by minc or multicast inference of network characteristics, have been developed to provide this ability. They consider multicast data packets to be probes, and conduct inference based upon reports of which probes reached each receiver. In practice, gathering reports from receivers in real time is a non-trivial task that presents scaling problems as the number of receivers increases. Prior work has led to an extension of the RTP data transport protocol to permit receivers to report per-probe information in packets known as RTCP XR packets.

This paper demonstrates how minc inference can, in fact, be conducted using only a default RTP packet format known as RTCP RR. RTCP RR packets contain summary information rather than per-probe information. They thus offer bandwidth savings, although this comes at the expense of an increase in estimator convergence time. Furthermore, this technique can be used by the observer of any standard RTP session, whereas estimation based upon per-probe information is only possible when a session explicitly employs the extended reporting format.

Keywords: End-to-end measurement; Moment estimator; Multicast; RTCP; Loss inference; MINC

Article Outline

1. Introduction

1.1. Loss inference from multicast probes

1.2. RTCP extensions for loss reporting

1.3. Contribution

1.4. Outline

2. Tree model and probe process

2.1. Multicast tree model

2.2. Probe model

3. The MINC loss estimator

3.1. Measurements and data

3.2. Two leaf tree

3.3. General tree

4. Model for RTCP reports

4.1. Reports and blocks

4.2. Statistical model of block sizes and counts

5. Moment estimator for two receivers

6. Moment based estimator for general trees

7. Estimation variance

8. Experiments

8.1. Alignment

8.2. Estimator combination

8.3. Link loss rates and randomization

8.4. Accuracy measures

8.5. Two receiver trees: dependence on block size

8.6. Two receiver trees: accuracy of variance approximation

8.7. Impact of report loss

8.8. Impact of multicast group joins and leaves

8.9. Inference on larger trees

9. Conclusions and further work

References

Vitae