[55576] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

Re: shuttle flash crowd statistics

daemon@ATHENA.MIT.EDU (Sean Donelan)
Sat Feb 1 18:05:28 2003

Date: Sat, 1 Feb 2003 18:02:35 -0500 (EST)
From: Sean Donelan <sean@donelan.com>
To: nanog@merit.edu
In-Reply-To: <20030201114551.E66266-100000@www.mauigateway.com>
Errors-To: owner-nanog-outgoing@merit.edu


On Sat, 1 Feb 2003, Scott Weeks wrote:
> BTW folks are interested, but there is little data coming in to share.
> As time goes on, I hope folks that show unusual traffic levels (on both
> sides; eyeball networks and content networks as well as transit networks)
> will send pointers to me that I can share with others.  I am very
> interested in flash crowd situations and how to mitigate the problems
> associated with them...

Historically providers have been reluctant to provide that level of
detail concerning traffic levels.  A few providers, generally smaller
ones, do make MRTG graphs available.  Once in a while a provider will
announce they had X Peta/Terrabytes of traffic for some time period.
But most prefer measurements which can not be correlated with revenue
(e.g. packet drops, latency, jitter, availability, etc).

I didn't see any noticable change on Abovenet/MFN's public MRTG graphs
at MAE-West/MIX-West.  Keynote/Matrix public data show no visible changes
of the web sites they measure.  They may have some private data which
shows more details.

Of course, NASA is a US federal agency, so you can always try the Freedom
of Information Act (FOIA).  But they have other things to do today.


home help back first fref pref prev next nref lref last post