Re: Source code for TCP over satellite

[Gorry Fairhurst <[e-mail]>]

It may also be worth looking at the RFCs produced by the 
Working Group to which this mailing list once belonged.
I suggest that working group did NOT conclude TCP didn't work
over long delay links - but did identify some specific 
performance issues.  The impact on short TCP transfers
was most significant, and one that web traffic suffers
if the browser insists on doing many short transfers
(this is by no means a necessity in the days of http 1.1).


2488 Enhancing TCP Over Satellite Channels using Standard Mechanisms.
     M. Allman, D. Glover, L. Sanchez. January 1999. (Format: TXT=47857
     bytes) (Also BCP0028) (Status: BEST CURRENT PRACTICE)

http://www.ietf.org/rfc/rfc2488

2760 Ongoing TCP Research Related to Satellites. M. Allman, S.
     Dawkins, D. Glover, J. Griner, D. Tran, T. Henderson, J. Heidemann,
     J. Touch, H. Kruse, S. Ostermann, K. Scott, J. Semke. February 2000.
     (Format: TXT=111141 bytes) (Status: INFORMATIONAL)

http://www.ietf.org/rfc/rfc2760

Sure, if you are talking of delays of the orders of 10s
of seconds and above - things look pretty horrible,
but most people don't experience this sort of delay.

Gorry Fairhurst

Ed Kraemer wrote:
> 
> This is true. The very things that help to make
> TCP/IP useful for things like the Internet, make
> it less useful over long-propagation-delay networks.
> 
> Here are two links that you may find helpful.
> 
> The second is a site that has lots of good papers
> relating to long propagation networks (space) and the
> issues therein. The first is a paper I wrote for a class
> that basically tries to sum up similar information.
> 
> Take care,
> Ed
> 
> http://students.cecs.csulb.edu/~ekraemer/
> http://www-scps.jpl.nasa.gov/scps/
> 
> > -----Original Message-----
> > From: [e-mail] [mailto:[e-mail]]On
> > Behalf Of Rajesh Raman
> > Sent: Wednesday, September 26, 2001 2:47 PM
> > To: [e-mail]
> > Subject: Source code for TCP over satellite
> >
> >
> > Hello,
> >
> > I have been led to believe that TCP does not perform
> > well over satellite.  Problems associated with it are
> > low throughput, throughput, etc.
> >
> > Could anybody tell me how these problems are solvable?
> >  Does anybody have products or source code that solve
> > these problems ?
> >
> > Thank you in anticipation.
> >
> > Rajeo
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger. http://im.yahoo.com