home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!fuug!demos!kiae!glas!demos!werple.pub.uu.oz.au!markd
- From: markd@werple.pub.uu.oz.au
- Newsgroups: comp.mail.uucp
- Date: 18 Jul 92 06:09 MDT
- Subject: Re: Maximum line latency
- Sender: Notesfile to Usenet Gateway <notes@glas.apc.org>
- Message-ID: <1992Jul18.020912.22633@werple.pu>
- Nf-ID: #N:1992Jul18.020912.22633@werple.pu:1773226347:001:1163
- Nf-From: werple.pub.uu.oz.au!markd Jul 18 06:09:00 1992
- Lines: 33
-
-
- ross@zooid.guild.org (Ross Ridge) writes:
-
- >markd@werple.pub.uu.oz.au (Mark Delany) writes:
- >>>I've heard rumors
- >>>of high latency satellite links and the like, but these are not
- >>>subjects I know anything about.
- >>
- >>You need to add at least a second for a satellite link.
-
- >I believe CCITT standards limit the delay to only a few tenths of
- >a second. Specfically I remember that one satallite hop was ok,
- >but two satallite hops would create too long of delay.
-
- Well, I suppose we should be precise about what we're measuring.
-
- The distance between receivers for a geo-stationary is ~ 75,000km
- giving ~ .25 of a second EM propogation time in one direction (the
- figures are too rough to worry about medium).
-
- So, on a perfect satellite link with no echo suppression getting in
- the way and a small data packet, you might just get an ack back in
- fractionally over .5 of a second.
-
- In reality, your timers will have to be of the order of multiple
- seconds for a large window protocol anyway, so whether you can make
- any sensible allowance for a satellite link is probably moot.
-
-
-
- --
- Mark Delany markd@werple.pub.uu.oz.au
-
-