Re: [Nolug] Tulane

On Sat, 19 Feb 2005 14:23:35 -0600, Brett D. Estrade <estrabd@yahoo.com> wrote:
> Sounds shady to me...

I am glad to see people suspicious by default in this day and age.
However, I assure you, I do not want a shell account to HaX0r or DoS
someone's boxen.

> why not go contact Tulane directly?

Because most of the time, before doing that, it is often easier to ask
a group of techies if they know someone that works in the IT
department who would know by default where/who I should go to.

> I am sure they
> would be more than happy to help.

You would be surprised what the general student worker or the typical
secretary, etc. at all of the published "help desk" numbers on the web
page would respond with.

> Also, even if this is a legit
> request, you should know that accessing only the Abilene network
> directly gives you the OC3 bandwidth you want. Accessing it through
> LaNet only gives you access to the network, not its raw bandwidth.
>

Umm, technically, LaNet is also the pipe that leads into the Abilene
network. LaNet plugs into the Houston Abilene router. Tulane has
their own pipe into that router. View these pages:

http://loadrunner.uits.iu.edu/weathermaps/abilene/ (click on the
houston router, scroll down till you see LaNet and/or the Tulane pipe.
 Also, Abilene is a 10 gig [OC192] pipe, so no matter what, I will not
see the "raw bandwidth" of Abilene.)

http://www.internet2.edu/resources/AbileneMap.pdf (look at the Houston
router. You will notice an OC3 leading to LSU named LaNet and another
leading to Tulane.)

A similar map to the above PDF: http://www.abilene.iu.edu/doc/logical.html

Let me explain a bit more. My hope was to find someone at Tulane that
could run a test for me. The reasoning behind this is that Tulane and
LSU both tie into the same Abilene router in Houston. Literally
inches apart in a Juniper networks T640 router as seen in this photo:
http://abilene.internet2.edu/images/AB-rack.jpg. Now, if I can find
someone at Tulane who can run a bandwidth test using a client for
which we run a daemon on a server at the observatory 24x7, it would
eliminate most points where congestion and/or bottlenecks could
happen. Assuming I can find someone at Tulane, then the bandwidth
test would essentially look like this: Unknown Tulane LAN hardware
-->Tulane Border Router --> Tulane OC3 Abilene link --> Core router at
Houston --> OC3 LaNet --> LSU Border router --> gigabit LAN --> LIGO
Router 1Gbps --> server. This would essentially be a 6 hop network
with the slowest links being OC3s eliminating most of the things that
could be causing a bottleneck to Caltech. The elimination of as many
possible points is what I am after.

I also have a call into the Abilene NOC at IU. They may have a box
hanging off of the Houston router that would eliminate even more hops,
but until I hear from them I am also interested in testing from
Tulane.

> On Fri, 18 Feb 2005 19:24:23 -0600, "Shannon Roddy" <sroddy@gmail.com>
> said:
> > Hi All!
> >
> > First, I should introduce myself. I am a Linux user since ~1995 from
> > Baton Rouge and a regular on brlug.
> >
> > Second, does anyone here have a shell account on a Tulane machine that
> > can run a bandwidth test for me? Or know someone at Tulane? We peer
> > into Internet2 via LaNet through LSU and we are having some "issues"
> > with our bandwidth back to Caltech from the observatory here in
> > Livingston (www.ligo-la.caltech.edu). It appears that Tulane has its
> > own OC3 into the Houston Abilene router and would be an ideal location
> > to run a test from.
> >
> > Thanks,
> > Shannon Roddy
> > ___________________
> > Nolug mailing list
> > nolug@nolug.org
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> http://www.brettsbsd.net/~estrabd
>
> ___________________
> Nolug mailing list
> nolug@nolug.org
>
___________________
Nolug mailing list
nolug@nolug.org
Received on 02/20/05