Re: Trace selection, schedules and duration

New Message Reply About this list Date view Thread view Subject view Author view

From: Joerg Micheel (joerg@cs.waikato.ac.nz)
Date: Mon Jan 22 2001 - 14:13:05 PST


Bo,

On Mon, Jan 22, 2001 at 11:32:11AM -0800, Bo Ryu wrote:
> How powerful are these real-time trace monitors/collectors? For example, can
> they do more than timestamping packets and storing packet headers? Such as
> additions, multiplications with floating point numbers, etc? Or can they
> compute flow-level data on the fly at OC3 speed or higher?
>
> Am I way out of mind? :-)

As much as I am. I think this kind of technology has to be developed,
traces are not the answer to everything (although to a lot of questions
already). I think it is possible to achive software-based realtime
analysis into the hundreds-of-megabits link load. For gigabits, this
looks a bit unrealistic, but we have the support of hardware in this
case. Modern OC48c monitoring gear has a hell of a lot of extra hard-
ware onboard to cater for our needs, if only we knew what the needs
are ...

We'll have to do realtime evaluation of algorithms for data analysis,
such as flow state engines, packet statistics etc.

        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


New Message Reply About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b30 : Thu Sep 27 2001 - 16:24:41 PDT