From rusty at rustcorp.com.au Fri Sep 25 01:53:42 2015 From: rusty at rustcorp.com.au (Rusty Russell) Date: Fri, 25 Sep 2015 11:23:42 +0930 Subject: [Lightning-dev] Ionization Protocol: Flood Routing In-Reply-To: References: <874mioecuq.fsf@rustcorp.com.au> Message-ID: <87y4fvqn6h.fsf@rustcorp.com.au> Amos Bairn writes: > The obvious problem with having the whole network use the same set of > beacons at the same time is that those nodes will get slammed. A possible > solution is to have each user pick their own set of beacons (at random), > but make the beacon set large enough so that any two users are likely to > share a few. That way the all of nodes would share the load roughly evenly. The core scaling problem is propagating current fee information (ie. the dynamic part of route information) across the network. That's easy with agreed beacons. Hard with random beacons. Beacons are going to get slammed, but it's not clear how bad it is. Getting slammed from all directions doesn't use up channels; it's only computational and bandwidth limits. Their neighbors will want to buff up, too (they'll take some load off the beacon if both parties route through them). A semi-realistic simulation would be interesting. If payments cluster by geography and some random channels are established by proximity (ie. low latency) that may take some of load off the beacons too. Cheers, Rusty.