Pushing more than 936kb/sec via satellite

Hank Nussbacher <hank@ibm.net.il> Sun, 15 November 1998 14:30 UTC

X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcpsat@lerc.nasa.gov using -f
Message-Id: <2.2.32.19981115143027.006bdb48@max.ibm.net.il>
X-Sender: hank@max.ibm.net.il
X-Mailer: Windows Eudora Pro Version 2.2 (32)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sun, 15 Nov 1998 16:30:27 +0200
To: tcpsat@lerc.nasa.gov
From: Hank Nussbacher <hank@ibm.net.il>
Subject: Pushing more than 936kb/sec via satellite
Sender: owner-tcpsat@lerc.nasa.gov
Precedence: bulk
Status: RO
Content-Length: 854
Lines: 19

I have read over:
http://www.ietf.org/internet-drafts/draft-ietf-tcpsat-stand-mech-06.txt
and have some questions.  I am not on your list, so please remember to reply
also directly to me.

Today, is there any way without modifying the TCP stacks in NT, W95, AIX,
Solaris, SG, HP, etc. to push more than 936kb/sec via satellite?  Can one
put a black box on each end and fake things out (window size, ACKs, NAKs,
etc.) for all TCP streams so as to get 20-30Mb/sec per TCP stream via GEO
satellites?  Does anyone know anyone doing this?  Does anyone know of
equipment that can handle T3/OC-3 speeds?

What if the GEO satellite link was simplex with a fiber uplink?  This would
alter the numbers - has anyone had any experience with hybrid systems being
able to pump 20-40Mb/sec per single TCP stream?

Thanks in advance for any info,
Hank Nussbacher
Israel

Pushing more than 936kb/sec via satellite Hank Nussbacher
Re: Pushing more than 936kb/sec via satellite Brad Smith
Re: Pushing more than 936kb/sec via satellite Anil Agarwal