On Wed, Dec 13, 2006 at 06:51:55PM -0500, Juiceman wrote: > On 12/13/06, toad <toad at amphibian.dyndns.org> wrote: > > On Wed, Dec 13, 2006 at 03:14:39PM +0000, Michael Rogers wrote: > > > toad wrote: > > > > This is why the vast majority of swap attempts fail in > > > > "swapsRejectedAlreadyLocked" on the stats page: > > > > > > If most nodes are locked at any given time, doesn't that mean nodes are > > > spending most of their time executing swaps, ie swaps are already > > > happening nearly as fast as they can? Sorry if I've misunderstood. > > > > Most of the time they are trying to send swaps out, which get rejected > > because the node they reach is already locked, because it itself is > > sending a swap out. Executing swaps doesn't take long; it's the 6 hops > > random routing to find a partner who isn't already locked, the reply, > > the confirmation and the completion that take time; but maybe we can > > significantly speed this up by queueing accepts, so that nodes don't > > fail to swap most times? > > Forgive me if I misunderstood the swapping algo or what you are > proposing, but how can we queue accepts when our node location is > about to change? Won't that affect who we should swap with next?
The selection of who to try to swap with is random. We can't accept the request and send our location and that of our neighbours until current and queued swaps have completed for that reason. -------------- 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/devl/attachments/20061214/1e14155e/attachment.pgp>
