Did you compare mandatory with advisory backoff? A few months ago we concluded that mandatory backoff wasn't worthwhile because we have a separate mechanism (client-side load limiting) to limit load, and also that it wasn't fair on users with few connections, and so made backoff advisory rather than mandatory. I.e. a node will consider nodes which are not backed off first, but will route to backed off nodes if it has no other alternative.
Obviously the truth will be best known if a minority of the simulation are very slow / overloaded by external causes, as I said before. On Wed, Nov 22, 2006 at 12:11:44AM +0000, Michael Rogers wrote: > Here are some preliminary results from the simulator - I must stress > that they're only preliminary. I haven't simulated token passing yet - > these results only show throttling with backoff, throttling alone, and > backoff alone. > > The load model is a bit simplistic: one in ten nodes is a publisher, and > each publisher has ten randomly selected readers. Each publisher > occasionally inserts a key, waits for ten minutes, then informs its > readers of the key; the readers then request the key. The publication > rate (and therefore the request rate) can be varied to investigate the > effect of load. > > Each run lasted for three hours' simulation time, with the first hour's > logs discarded to minimise the effect of the initial conditions. > > All three mechanisms showed an increase in throughput under increasing > load, ie there was no congestion collapse. Throttling alone produced > higher throughput than either throttling with backoff or backoff alone, > especially under heavy load. > > All three mechanisms showed a decrease in success rate with increasing > load, suggesting that congestion collapse might eventually occur at high > enough loads. Throttling alone produced a higher success rate and slower > degradation under load than either throttling with backoff or backoff alone. > > This suggests that the backoff mechanism is not effective in controlling > load, and the request throttle would work better without backoff. These > conclusions are only tentative though - much more remains to be done, > when I can find enough disk space for the logs! > > Cheers, > Michael > _______________________________________________ > Tech mailing list > Tech at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/tech -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/tech/attachments/20061122/212f414f/attachment.pgp>
