(no subject)

This paper has discussed findings from a large-scale study of
Internet packet dynamics conducted by tracing TCP bulk transfers
between 35 Internet sites. The measurement allow to distinguish
between the end-to-end behaviors due to the different directions of
the Internet paths. This paper mainly reports on a large-scale
experiment to study end-to-end Internet packet dynamics.

The paper characterizes the prevalence of unusual network events
including out-of-order delivery, packet replication, and packet
corruption according to the trace. Then authors discuss a robust
receiver-based algorithm for estimating "bottleneck bandwidth". The
bottleneck estimation technique used in previous work is based on
"packet pair". However, there are some fundamental problem in "packet
pair" method, including Out-of-order delivery, Limitations due to
clock resolution, Changes in bottleneck bandwidth, and Multi-channel
bottleneck links. Motivated by the shortcomings of packet pair, the
authors developed a significantly more robust procedure, "packet
bunch modes". The main observation behind PBM is that authors can
deal with packet-pair's shortcomings by forming estimates for a range
of packet bunch sizes, and by allowing for multiple bottlenect values
or apparent bottleneck values.

After that, authors investigated patterns of packet loss, finding
that loss events are not well-modeled as independent and,
furthermore, that the distribution of the duration of loss events
exhibits infinite variance; and analyzed variations in packet transit
delays as indicators of congestion periods, finding that congestion
periods also span a wide range of time scales.

This paper is well written and presented. The contributions are based
on the large-scale tracing. Moreover, with due diligence to remove
packet filter errors and TCP effects, TCP-based measurement provides
a viable means for assessing end-to-end packet dynamics. However, I
wonder if the data collected in the paper is enough for the analysis.
Received on Tue Nov 14 2006 - 23:30:03 EST