[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference noted::atm

Title:	atm

Moderator:	NPSS::WATERS

Created:	Mon Oct 05 1992
Last Modified:	Thu Jun 05 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	970
Total number of notes:	3630

960.0. "questions raised by customer pl. answer" by XFMV01::DIVYESH () Fri May 30 1997 02:42

    Hi,
    
      At my customer site BARC i have implemented CLIP over SVC. This
    cutomer has certain queries which are as follows.
    
    1) What are smart PVC's ?
    
    2) How much time VC signalling takes to eshtablish SVC between two
       systems ?
    
    3) Does SVC's times out if there is no network traffic and if so then
       how to increase that time out value ?
    
    4) Is network implemented in ELAN is slower then in CLIP ?
    
    Thanks in advance.
    
    
    Divyesh

T.R Title User Personal
Name Date Lines

960.1 NPSS::NEWTON Thomas Newton Fri May 30 1997 06:33 15

T.R	Title	User	Personal Name	Date	Lines
960.1		NPSS::NEWTON	Thomas Newton	`Fri May 30 1997 06:33`	15
	> 1) What are smart PVC's? I think smart PVCs are some sort of PVC/SVC hybrid. They look like PVCs to the end nodes, in the sense that they appear to be "permanent", but they're actually implemented using some sort of signalling. The idea is that if some node in the network goes down, disrupting the VCC, the network nodes connected to the endstations can detect this and set up a new VCC (maybe through regular signalling, maybe through standby circuits - I'm not sure). With a regular PVC, a prolonged switch failure would result in prolonged service outage ... someone would have to notice the dead node, then go in through management to reroute the circuit. This is all based on something I heard long ago, so it could be fuzzy.

>   1) What are smart PVC's?

I think smart PVCs are some sort of PVC/SVC hybrid.  They look like PVCs to
the end nodes, in the sense that they appear to be "permanent", but they're
actually implemented using some sort of signalling.

The idea is that if some node in the network goes down, disrupting the VCC,
the network nodes connected to the endstations can detect this and set up a
new VCC (maybe through regular signalling, maybe through standby circuits -
I'm not sure).  With a regular PVC, a prolonged switch failure would result
in prolonged service outage ... someone would have to notice the dead node,
then go in through management to reroute the circuit.

This is all based on something I heard long ago, so it could be fuzzy.