HP3000-L Archives

April 1997, Week 5

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jeff Kell <[log in to unmask]>
Reply To:
Jeff Kell <[log in to unmask]>
Date:
Tue, 29 Apr 1997 17:39:47 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
Gavin Scott wrote:
>
> Wirt wrote:
> > The "LAN-based" point-to-point system that Jeff describes above
> > re-invents statistical multiplexers (if these new devices run a
> > common synchronous clock between them and commutate, and thus
> > maximally pack, the packets).
>
> I think ATM is a better example of something that works this way.

And ATM is just Frame Relay with fixed length frames.

Back to full-duplex 100TX/FX... when working point-to-point, it
effectively removes "carrier sense" and "collision detection" from the
loop -- you just talk when you want.  You have to be careful, however,
when using old "dumb" shared media hubs; you generally can't do full
duplex on them.

cisco recently announced some new technology (ether channel?) where you
can assign 1, 2, 4, or 8 100Mb full-duplex channels together and the
devices either bit-slice or byte-slice (or something) so that you get
aggregate throughput that is actually 2x, 4x, or 8x the single link.
It is not a "load sharing" but actually transmitting in parallel, so
you can get (in theory) an 800Mb full-duplex "pipe" using 8 pairs.

Jeff Kell <[log in to unmask]>

ATOM RSS1 RSS2