Hi, Bruce,
From: "Bruce Lowekamp" <[EMAIL PROTECTED]>
On Jan 10, 2008 3:17 AM, Pekka Nikander <[EMAIL PROTECTED]>
wrote:
Hi Bruce,
> The hip routing table approach looks great from a layering
> perspective, but implementation looks to be very hard.
Why? Do you allude that the various overlay routing protocols are
sufficiently different from the IP routing protocols so that one
cannot compute a forwarding table? Or is there some other
implementation problem that you are thinking about?
The issue I'm concerned about here is that different DHTs use
completely different routing algorithms. For example, for Chord you
forward the message to the peer with the closest ID that is < the
target, but with Kademlia, you do a prefix match and calculate
distances with xor. I'm sure it's possible to come up with a safe
language or something that could be used to describe how to make
routing decisions, but it's certainly non-trivial.
This is my concern also, FWIW. I also think it's possible, and I also think
it's non-trivial.
> So in the hip routing table approach, the I1 etc messages are
> routed via 5 using HIP to 10 and eventually a new direct connection
> is established.
Well, depending on many factors, it may be only a subset of the HIP
base exchange messages that would be routed over the overlay (through
node 5); some of them (like perhaps R2) could perhaps be sent
directly, at least in some cases. So, there is some flexibility
here, but at least I don't quite understand the situation in cases
where both 1 and 10 are behind NATs.
As long as they are both reachable in the overlay, both peers behind
nats isn't a problem as long as the ice message exchanges are carried
over the overlay.
I'm not saying there is a chicken-and-egg problem, but I'm not sure that
there isn't (yet).
Thanks,
Spencer
_______________________________________________
P2PSIP mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/p2psip
