Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: Renaming the WAN PHY




Nevin and Gary,

I agree with both of you and see another problem with "SONET lite".

When we convert "SONET lite" into "SONET-compliant" at the edge of 
SONET infrastructure, we require ELTE (Ethernet Line Terminating 
Equipment) to re-clock with +/-4.6ppm accuracy.  This means we 
should re-write the pointer for the stuff/destuff; this is almost 
the same task as SONET framing as Mr. Paul Bottorff has shown in 
his Albuquerque slide pages 17-18.
http://grouper.ieee.org/groups/802/3/ae/public/mar00/bottorff_2_0300.pdf

I don't believe we need "SONET lite" just for the jumper connection 
to ELTE, since we already have an elegant solution to provide 
OC-192c/VC-4-64c compatible MAC rate; see Mr. Shimon Muller's 
Albuquerque proposal.
http://grouper.ieee.org/groups/802/3/ae/public/mar00/muller_1_0300.pdf
I think this is enough to meet the PAR scope of 802.3ae.
http://grouper.ieee.org/groups/802/3/ae/PAR-802-3ae.pdf

I agree that SONET-compliant PHY makes sense since it can be 
connected directly to install-based SONET/DWDM network.  However, 
in this case at present I support Nortel's T1X1.5 EOS Proposal since 
it is the only complete solution to cover all SONET payload rates.  
Its less mapping overhead is also preferable. I don't want to see 
two different payload mapping in OC-192c.

These are the backgrounds of my 10GENIE&FFRC proposal in Albuquerque. 
Now I have realized that Shimon's open loop control is much better 
than FFRC, in Ottawa I will up-date my proposal by focusing on how 
10GENIE realizes the "SONET-(lite)-compliant OAM&P".  I believe that 
10GENIE is the most economical solution to date to meet the PAR 
purpose of 802.3ae; to expand the Ethernet application space to 
include WAN links.

Any comments for my 10GENIE proposal would be greatly appreciated.

Best Regards,
Osamu

---------------------------------------
Osamu Ishida
NTT Network Innovation Laboratories
TEL:+81-468-59-3263 FAX:+81-468-55-1282
---------------------------------------

At 6:03 PM -0500 00.3.26, Jones, Nevin R (Nevin) wrote:
> Gary:
> 
> Well put! You have hit on exactly what bothers me with with Dave's
> characterization of the WAN PHY as not having to be "SONET compliant".
> 
> I believe that whether the inferface is "SONET Lite" or "compatible" it is
> going to have to work with the existing deployed SONET interfaces.
> 
> It is precisely this sort of  fuzziness about SONET "lite" that caused its
> demise at the OIF last year.
> 
> -Nevin Jones 
> 
> > ----------
> > From: 	Gary Nicholl[SMTP:gnicholl@xxxxxxxxx]
> > Sent: 	Friday, March 24, 2000 5:59 PM
> > To: 	David Martin; stds-802-3-hssg@xxxxxxxx
> > Subject: 	RE: Renaming the WAN PHY
> > 
> > 
> > Dave,
> > 
> > Having been down a similar path with POS (Packet-over-SONET) I don't see
> > how
> > you can have a SONET compatible phy that isn't SONET compliant.  As I
> > understood it the reason for having a SONET framed interface in the first
> > place
> > was to be compatible with the installed SONET based transport and DWDM
> > networks. To meet this requirement I think the interface has to be SONET
> > compliant, in the same way that POS interfaces have to be compliant with
> > all
> > the appropriate Bellcore and ITU SONET/SDH standards.
> > 
> > I think it would be very confusing to  to have an interface that 'looks
> > and
> > feels' like SONET but in fact isn't. If that is really the intent then I
> > think
> > it needs to be clearly distinguished from a 'real' SONET interface.
> > 
> > Gary Nicholl .............