[c-nsp] T1 Bonding with PA-MC-T3

[Nick Voth]

> On Fri, 14 Mar 2008, Nick Voth wrote:
> 
>> and it seems understandable, but I'm not totally sure what it's doing yet.
>> There's per-packet and per-destination load-sharing and it's not really
>> clear what would be appropriate for my situation.
> 
> Both work well.   per-packet generally gives you the most even
> distribution between links, and enables you to "see" the aggregate
> throughput (in past experience, it has been a few years) in a single
> transfer.
> 
> If your two links traverse different paths, though, you have a risk of
> packets arriving at different times.   per-destination helps with this
> problem, by basically sending any one flow down only one link.  You don't
> get to realize the full bandwidth in a single transfer, though.   Any
> single transfer maxes out at the speed of the link.
> 
> 
> If your two T's travel alongside each other for their full path,
> per-packet usually works just fine.
> 
> It /does/ require having a device on the other end that supports it....but
> if it's a Cisco that can handle two T1 interfaces, that shouldn't be a
> problem.
> 
> 
> 
> ..david
> 
> 

Thanks very much David. That definitely helps. Yes, our 2 T's are on the
same path to the destination so it looks like per-packet would be best in
that case. However, with "per-packet" can you utilize the full speed of the
2 T's as if they were bonded like in MLPPP? That's the ultimate goal here.

Thanks,

-Nick Voth