Any chance of having distributed multilink PPP on regular T1/E1 interfaces,
if channelized T3 is not available?
I could use a "channelizer" but I think that would introduce an additional
point of failure?
So, according to the docs:
http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/120newft/120t/120t3/multippp.htm
"Non-distributed MLPPP can only perform limited links, with CPU utilization quickly reaching 90% with only a few T1/E1 lines running MLPPP. With distributed MLP, you can increase the router's total capacity."
Non-distributed meaning on a non-RSP platform?
The setup I have in mind is two sites in the US with 15 T1's each, going to
at least two sites here in the Philippines. I will have channelized T3 in
the US but not here.
So, another other option would be to run multiple loopback BGP sessions.
If CEF supports up to 6 parallel path for per-destination forwarding (right?),
then will 3 BGP sessions in parallel work? The idea is that we can use this
BGP to fine-tune what passes through each group of lines.
Will one VXR for each site do?
Am I right that it takes more resources to run 30 T1's in parallel that a
single DS3, at full rate?
--http://www.internet.org.ph Internet and ISP's in the Philippines http://www.ASARproject.org Artists for Social Action and Response
GSM Mobile: +63-917-810-9728
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:24 EDT