jamin@JERICO.USC.EDU (Sugih Jamin) (06/05/91)
The following paper, to appear in ACM SIGCOMM '91, is available via
anonymous FTP from jerico.usc.edu (128.125.51.6):
The file is ~ftp/pub/jamin/traffic/traffic.ps.Z Be sure to use binary
mode when doing the transfer.
__________________________________________________________________________
Characteristics of Wide-Area TCP/IP Conversations
Ramon Caceres+ Peter B. Danzig* Sugih Jamin* Danny J. Mitzel*
*Computer Science Department, University of Southern California,
Los Angeles, California 90089-0782
+Computer Science Division, University of California,
Berkeley, California 94720
traffic@excalibur.usc.edu
Abstract
In this paper, we characterize wide-area network applications that
use the TCP transport protocol. We also describe a new way to model
the wide-area traffic generated by a stub network. We believe the
traffic model presented here will be useful in studying congestion
control, routing algorithms, and other resource management schemes
for existing and future networks.
Our model is based on trace analysis of TCP/IP wide-area internetwork
traffic. We collected the TCP/IP packet headers of USC, UCB, and
Bellcore networks at the point they connect with their respective
regional access networks. We then wrote a handful of programs to
analyze the traces. Our model characterizes individual TCP conversations
by the distributions of: number of bytes transferred, duration, number of
packets transferred, packet size, and packet interarrival time.
Our trace analysis shows that both interactive and bulk transfer
traffic from all sites reflect a large number of short conversations.
Similarly, it shows that a very large percentage of traffic is
bidirectional, even for bulk transfer. We observed that interactive
applications send significantly different amounts of data in each
direction of a conversation, and that interarrival times for interactive
applications closely follow a constant plus exponential model. Half of
the conversations are directed to a handful of networks, but the other
half are directed to hundreds of networks. Many of these observations
contradict commonly held beliefs regarding wide-area traffic.
__________________________________________________________________________
Based on the workload characteristics reported in the paper, we would like
to build a wide-area workload model. To that end, we need to collect more
traces. If you are at a site on the Internet and are willing to assist us
in collecting wide-area traffic traces, please contact us at
traffic@excalibur.usc.edu.