Network analysis.

Hans-Werner Braun hwb at upeksa.sdsc.edu
Wed Dec 15 19:45:52 UTC 1993


Over the last few years we have spent a considerable amount of effort
at SDSC/UCSD to analyze traffic characteristics of wide area networks.
While we were able to just scratch the surface with what is doable and
useful, particularly for network planning, engineering, and research,
we are getting to a point where it would be useful for us to learn what
kind of statistics people currently collect, what they are doing with
those statistics, and what kind of statistical objects and analysis
they think would make their network work better (make it more
manageable, planable, or whatever).

So, we are interested in finding out for various operational
infrastructures in the U.S. as well as the international Internet
community, the following questions:


  (1) what kind of statistics do you collect?

  (2) how long of a history do you keep?

  (3) what is the volume of statistics being generated per day/week/month?

  (4) what are you doing with them in terms of post processing? Just
      for operational purposes? In depth analysis? Research, e.g., for
      published papers?

  (5) what kind of additional statistics and analysis would you find
      useful for your network?

  (6) anything else worth mentioning?


We would like to compile a list from the submitted responses, for
everyone's benefit. Specifically now with the beginning of much more
formalization of network engineering and operational groups that have a
strong emphasis on real infrastructure, it may be interesting to find
out what the statistics and network analysis requirements are for
operational environments, and whether some more of our work could be of
value to the Internet community. For example, now that the basic fabric
is workable, can we pull things together better in some way or another,
e.g., by having a better, and perhaps day-to-day operational,
understanding of international flows.

If you want more details about what we are doing with network analysis
and so, drop me a not and I will send you more information.

Hans-Werner





More information about the NANOG mailing list