Re: Two things about new firewalls etc.

> Seriously though, I think the paper is entirely relevent in data traffic
> pushed on some types of CSMA/CD (shared) mediums. Ethernet by nature is an
> unfair and random layer-1|2 protocol. However, it is not clear that the
> low-frequency components necessary for long-range dependence are present in
> token-based, ring or switched systems.

There has been followup work that strongly suggests self-similarity
arises from properties of traffic *sources*, in which case it doesn't
matter what kind of media it traverses. (If the network uses admission
control and rate shaping, then the source characteristics manifest themselves
in making efficient admission control a difficult problem.)

In particular:

        Self-Similarity Through High-Variability: Statistical Analysis of
                Ethernet LAN Traffic at the Source Level
        Walter Willinger, Murad S. Taqqu, Robert Sherman, Daniel V. Wilson

        IEEE/ACM Transactions on Networking, Vol. 5, No. 1, pp. 71-86, Feb 1997
        http://math.bu.edu/people/murad/pub/source-printed-version-posted.ps

and, for discussions of self-similarity in Internet traffic:

        Wide-Area Traffic: The Failure of Poisson Modeling
        V. Paxson and S. Floyd
        IEEE/ACM Transactions on Networking, 3(3), pp. 226-244, June 1995.
        ftp://ftp.ee.lbl.gov/papers/WAN-poisson.ps.Z

        Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes
        Mark E. Crovella and Azer Bestavros
        Proc. SIGMETRICS 1996
        http://www.cs.bu.edu/faculty/crovella/paper-archive/self-sim/
                sigmetrics-version.ps

- Vern
Received on Dec 03 1997