RE: [RPRWG] RFC 2892; The Cisco SRP MAC Layer Protocol - butdiscussing masterless ring
Dear Ray,
I agree w/ Khale, Fredrick, and Angela's reply for a different
reason. Several employers ago, I built token ring, token bus, and
FDDI products along w/ Ethernet. Having dynamic master is both
implementation and network management challenge. Fixed master
schemes (any form of polling) works better, but still does not have
the scalability nor spatial reuse of RPR. I rather not deal w/ Ring
master schemes if we could (and if some of the members want to
pursue it, don't -- I do not want to go through the efforts of
digging through past history to convince them otherwise :)). Any of
the proposed schemes, especially schemes w/ production shipments, is
far more desirable (in many of our metrics, scalability, BW efficiency,
plug-&-play, fault tolerance, etc) than any of the popular past token
passing schemes (and by the way, these past token schemes were invented
by best minds of IBM, DEC, etc, and I am not hopeful that we could do
better in the time frame we agreed).
regards,
Yong.
============================================
Yongbum "Yong" Kim Direct (408)922-7502
Technical Director Mobile (408)887-1058
3151 Zanker Road Fax (408)922-7530
San Jose, CA 95134 Main (408)501-7800
ybkim@xxxxxxxxxxxx www.broadcom.com
============================================
-----Original Message-----
From: owner-stds-802-17@xxxxxxxx [mailto:owner-stds-802-17@xxxxxxxx]On
Behalf Of Khaled Amer
Sent: Wednesday, April 04, 2001 8:29 AM
To: Ray Zeisz; dbrown@xxxxxxxxxx; stds-802-17@xxxxxxxx
Subject: Re: [RPRWG] RFC 2892; The Cisco SRP MAC Layer Protocol
Most of us are against it and prefer a purely distributed environment.
Khaled Amer
President, AmerNet Inc.
Architecture Analysis and Performance Modeling Specialists
Address: 13711 Solitaire Way, Irvine, CA 92620
Phone: (949)552-1114 Fax: (949)552-1116
e-mail: khaledamer@xxxxxxx
Web: www.performancemodeling.com
----- Original Message -----
From: Ray Zeisz <Zeisz@xxxxxxxx>
To: <dbrown@xxxxxxxxxx>; <stds-802-17@xxxxxxxx>
Sent: Wednesday, April 04, 2001 6:53 AM
Subject: RE: [RPRWG] RFC 2892; The Cisco SRP MAC Layer Protocol
>
> Do the .17 objectives prevent the election of a master? Clearly requiring
a
> specific node to be on the ring and be the master would not be a good
design
> point; however, it seems that having a process like 802.5's Active/Standby
> monitor election may not be a terrible thing to have.
> Has there been any direction set on this?
>
>
> Ray Zeisz
> Technology Advisor
> LVL7 Systems
> http://www.LVL7.com
> (919) 865-2735
>
>
>
>
> -----Original Message-----
> From: dbrown@xxxxxxxxxx [mailto:dbrown@xxxxxxxxxx]
> Sent: Thursday, March 29, 2001 4:54 PM
> To: stds-802-17@xxxxxxxx
> Subject: [RPRWG] RFC 2892; The Cisco SRP MAC Layer Protocol
>
>
>
> After reading through RFC2892, it is not clear to me how each node
> "discovers" which interface is on which ring (i.e. inner or outer).
>
> I'm not suggesting we follow this implementation, but I am curious how the
> assignment is done.
>
> It could be done with a master device, but I believe that goes against
one
> of our objectives, I'd know if they were posted:)
>
> It could be hard coded, but that might limit flexibility.
>
> Regards,
>
> Dave Brown
> Chief Architect
> MOSAID Semiconductor
> (613) 599-9539