On the characteristics and origins of internet flow rates

By Yin Zhang, L. Breslau, Vern Paxson, Scott Shenker. Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2002

427 citations recorded by the source corpus.

This paper considers the distribution of the rates at which flows transmit data, and the causes of these rates. First, using packet level traces from several Internet links, and summary flow statistics from an ISP backbone, we examine Internet flow rates and the relationship between the rate and other flow characteristics such as size and duration. We find, as have others, that while the distribution of flow rates is skewed, it is not as highly skewed as the distribution of flow sizes. We also find that for large flows the size and rate are highly correlated. Second, we attempt to determine the cause of the rates at which flows transmit data by developing a tool, T-RAT, to analyze packet-level TCP dynamics. In our traces, the most frequent causes appear to be network congestion and receiver window limits.

Subfields: Artificial Intelligence, Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/3a5dfae6a6decc15698c7962562646d99d64293a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

Citations in the corpus

Dynamic load balancing without packet reordering (2007, background): “However, the common limitations pointed to by all reviewers lie around the practical issues of such an approach.”
SketchVisor: Robust Network Measurement for Software Packet Processing (2017, background): “• T is approximated as a low-rank matrix: As network traffic is dominated by large flows [54, 59], few counters in T have much different values from other counters that are only accessed by small flows.”
A first look at traffic on smartphones (2010, methodology): “Using these datasets, we shed light on several aspects of smartphone traf.c.”
Characterizing radio resource allocation for 3G networks (2010, result): “First, similar to previous measurements of wired network using passive trace [33, 26], the finite duration (3 hours) may influence distributions of state machine characteristics.”
On dominant characteristics of residential broadband internet traffic (2009, background): “[56] analyzed Internet flow traces from various access, peering, and regional links within a Tier-1 provider in 2002 to understand from where performance bottlenecks arose.”
An in-depth study of LTE: effect of network protocol and application behavior on performance (2013, methodology): “Previous measurements for wired networks also suggest that for Internet ﬂows, there exist correlations among their size, duration, and rate [36, 22].”