From: Joerg Micheel (joerg@cs.waikato.ac.nz)
Date: Mon Jan 22 2001 - 14:04:47 PST
On Mon, Jan 22, 2001 at 10:10:39AM -0800, k claffy wrote:
> On Sat, Jan 20, 2001 at 07:15:27PM -0800, Neil Spring wrote:
> > Ok. Longer traces. I think we have the necessary cycles to do
> > filtering on traces, the more since it reduces disk IO, which
> > I see becoming scarce. I wonder whether it is practical from
> > an analysis point of view.
>
> Yes, the disk IO tradeoff makes sampling more attractive
> from a technical point of view. I am also curious to hear
> what sort of anaylses this scheme might prevent.
>
> any intra-flow characteristics,
> e.g., iat burstiness.
We ("the capture people") also have to think in practical terms. Collecting
long traces is hard work, although I cannot believe I am saying this right
now, but it is true. The amount of data shuffling and postprocessing,
ensuring the time is right, the disk space is there, the GPS works etc.
I rather give it a good shot and then cater for the need of a large group
of people. Filtering will ultimately cut the user base into a small fraction.
Just for your information I would like to mention that there are several
long traces available for your use. Auckland-II contains about half a
dozen 24 hour traces (in handy GB chunks), the NZIX-II data is a continuous
five day trace at a busy New Zealand peering point. I am not saying this
is representative of anything, just that this should get most of you
started.
http://moat.nlanr.net/Traces/Kiwitraces
With first results from your analysis we might have some hooks as to
what the next long traces should look like.
Joerg
-- Joerg B. Micheel Email: <joerg@cs.waikato.ac.nz> WAND and NLANR MOAT Email: <joerg@nlanr.net> The University of Waikato, CompScience Phone: +64 7 8384794 Private Bag 3105 Fax: +64 7 8585095 Hamilton, New Zealand Plan: PMA, TINE and the DAG's
This archive was generated by hypermail 2b30 : Thu Sep 27 2001 - 16:24:41 PDT