From: Edet Nkposong (edetn@microsoft.com)
Date: Tue Jun 29 2004 - 10:38:02 GMT-3
Hello,
Maybe I don't understand your response but PNNI uses the Dijkstra
Algorithm
To compute the shortest path so why wont it re-route when a better path
that
Satisfies the same QoS (using CAC) requirements is available.
Regards, Edet
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Joe Chang
Sent: Tuesday, June 29, 2004 1:52 PM
To: gladston@br.ibm.com; ccielab@groupstudy.com
Subject: Re: ATM SVC
My understanding is that ATM's PNNI is not a best path routing protocol
in
the same sense as RIP or OSPF. So long as the nodes can maintain
established
QoS for a VC the switching path will go unchanged. Compare with MPLS
which
would make your topology switch back to a shorter path upon recovery of
the
direct link.
----- Original Message -----
From: <gladston@br.ibm.com>
To: <ccielab@groupstudy.com>
Sent: Monday, June 28, 2004 11:29 AM
Subject: ATM SVC
> Does the LS1010 has the capability to discover that there is a better
path
between two points and reroute the call?
>
> This it the topology:
>
> RA------LS1010-1----------------------LS1010-2-----RB
> | |
> | |
> +------------LS1010-3--------+
>
> LS1010-1 can reach LS1010-2 directly or through LS1010-3.
> If the direct path goes down, LS1010-1 constructs the path through
LS1010-3.
> But when the directly path goes up again, the path keep using
LS1010-3.
> Is it a limitation of the PNNI?
>
> Thanks for any comments.
>
>
This archive was generated by hypermail 2.1.4 : Sat Jul 03 2004 - 19:40:52 GMT-3