RE: [PWE3] PWE3 question

[neil.2.harrison@bt.com]

Title: RE: [PWE3] PWE3 question

Zheng...I'd be happy to agree with you provided it broke nothing else.  But I have explained on the list several times that it is no good saying something is scalable in 1-dimension if it is broken in others.  mp2p LSPs create very hard fault management problems and they are not good constructs on which to offer VPN services IMO.  VPN services need more than isolation wrt addressing, they also need strong fault/performance isolation from each other.  Like-DS-class merging across VPNs on mp2p E-LSPs does not provide this per VPN fault/performance isolation.

 

regards, Neil

-----Original Message-----
From: Zheng Wang [mailto:zheng@teradiant.com]
Sent: 20 May 2002 23:45
To: David Allan; erosen
Cc: pwe3
Subject: RE: [PWE3] PWE3 question

 

Zheng:

I consider state and fabric to be equivalent. Each LSH involves some state at the LSRs in either end, therefore the cumulative number of LSHs in the network is indicative of the amount of state involved.

Dave;

The totoal number of LSHs is not an important metric - it is the number of state you keep in each box that matters. In other word, the max size of the table in your router is the issue here.

Cheers

Zheng