RE: QoS in Shared Media

["Naidu, Venkata" <Venkata.Naidu@Marconi.com>]

Hi

  Thanks for all those who replied and gave good pointers.

  Fred/David, Some one pointed me to ISSLL WG with this regard
  (I thought this WG is also off along with INTSERV & RSVP)

  Esp about RFCs 2814/2815/2816 - SBM (Subnet Bandwidth Manager)
  Integrated Services Over Shared and Switched IEEE 802 LAN 
  Technologies. I didn't go through these drafts completely 
  but, aren't they doing in the direction we are looking for...

  my question is, why not RSVP for MPLS not considering these 
  work. Recently David was asking "why RSVP doesn't have strict
  delay and delay variation control parameters like ATM?"
  I found some of the RFCs in ISSLL (2379, 2380, 2381 & 2382)
  have answer for that - Inserv has the answer for QoS in any media!
  Are these something different when compared to RSVP for LSPs?

  I am searching for what I missed?! :-)

-Venkata Naidu  


-> -----Original Message-----
-> From: Fred Baker [mailto:fred@cisco.com]
-> Sent: Friday, May 04, 2001 1:48 PM
-> To: Naidu, Venkata
-> Cc: 'mpls@uu.net'; 'te-wg@ops.ietf.org'
-> Subject: Re: QoS in Shared Media
-> 
-> 
-> At 10:58 AM 5/4/2001 -0400, Naidu, Venkata wrote:
-> >I have a doubt regarding "QoS in Shared Media".
-> 
-> many have doubts in this area.
-> 
-> Fundamentally, if I can talk with a switch, I am in control 
-> of my end of 
-> the line and it is in control of its end, and we can make 
-> promises about 
-> what we will each do in our direction. On shared media, 
-> anything resembling 
-> QoS assertions or guarantees requires making promises on 
-> behalf of other 
-> systems one does not control. The nearest solutions I see to 
-> that have been 
-> in 802.5 and FDDI LANs, and I wouldn't describe those as having been 
-> "wildly successful."
->