[creswick at gmail.com: Re: [PLUG] Transcontinental ping times? for distributed symphonies]

Keith Lofstrom keithl at kl-ic.com
Wed Sep 8 13:03:01 UTC 2004


----- Forwarded message from Rogan Creswick <creswick at gmail.com> -----

Forward this to plug if you want, I can't send from my subscribed
address at the moment.

Very cool idea!

>From IBM TJ Watson in Hawthorne, NY (just north of Manhattan) I get
~83ms pings to Oregon State University, in Corvallis.  Pinging
uoregon.edu (which I think serves OSU's network) shaves about 2 ms off
that time, so 80-81 ms on average.

Targeting ucsd.edu (San Diego) comes in at about 86-87 on average, but
the median is 85.  This came as a bit of a surprise to me, I hadn't
expected the network to be so fast (or direct, or whatever is causing
only a 5 ms difference accross 1000 miles) in SoCal compared to
Oregon.  The stdev. is higher going to ucsd though. (about 2.5)

-Rogan

On Wed, 8 Sep 2004 10:22:15 -0700, Keith Lofstrom <keithl at kl-ic.com> wrote:
> 
> Question:  With fast machines at each end, and fat pipes running close
> to the main internet hubs, what is the fastest ping times seen running
> from coast to coast?

--- end of forwarded message.

Keith adds a small note:  NYC to PDX probably uses fiber links routed
through Seattle.  There is very likely another link going directly
from NYC to LA  ( a "traceroute" would tell you ).  The straight line
distances are within a few percent of each other.

This is demonstrated by air travel - from Portland to Boston is about
8 hours travel time, stopping at a hub, while direct flights from Boston
to LAX take 6 hours in fast widebodies.  Direct routes have advantages.

Keith

-- 
Keith Lofstrom           keithl at ieee.org         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs




More information about the PLUG mailing list