Cell Relay Archive

Cell Relay Retreat>List Archive>month:1997-Feb> msg00259



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Re: 802.3/802.5 Transparent Bridge Mis-sequencing over ATM

  • From: Ajay Sood -B7317 -Systems -GE <sood@genius.tisl.soft.net>
  • Date: Thu, 27 Feb 1997 09:24:21 +0500 (PAKST)

On 26 Feb 1997, jgrant wrote:

> 
> Distribution: world
> 
> At 23:37 25/02/97 GMT, Kris wrote:
> >Do all ATM networks shuffle the sequence of frames across the network?
> >I know that SMDS networks have parallel paths which can develop race
> >conditions so that frames between two CPEs can arrive in
> >non-deterministic order.
> 
> ATM networks can't shuffle cells on a virtual circuit (except 
> very occasionally under fault conditions) because then AAL5 
> wouldn't work. Hence it can't shuffle frames either, but only 
> on a single virtual circuit.
> 
> There are no guarantees about traffic on different virtual 
> circuits, which could go by quite different routes.
> 
> I haven't looked closely enough at LAN emulation to know whether 
> when it sets up a virtual circuit the first packets can overtake 
> packets that have already been sent via the BUS (Broadcast and
> Unknown Server).
> 
> 
> John Grant                    ______________________
> Director                     /                      |
> K-NET R&D Ltd               /      K  -  N  E  T    |
> +44 1223 862125            /________________________|
> http://www.k-net.co.uk     Leaders in ATM technology
> 
>