Re: LIFO simulations

Matthew Toseland <toad-EI5O+8PHWbJeeLb3ft/[email protected]> Fri, 15 Dec 2006 19:54:34 +0000

Newsgroups	gmane.network.freenet.technical
Message-ID	<[email protected]>

On Fri, Dec 15, 2006 at 03:24:02PM +0000, Michael Rogers wrote:
> Ken Snider wrote:
> > Couldn't you just accomplish this with a FIFO queue of arbitrarily short 
> > depth or fixed lifetime per-queue-object? The would seem to gain you all the 
> > benefits of LIFO without the drawbacks mentioned by toad.
> 
> A short FIFO queue for incoming searches would be similar to 
> token-passing, but tokens have the advantage that you know whether your 
> search will fit in the queue (because you have a token), so you can 
> avoid sending searches that will be dropped.

Right. A short FIFO queue for searches, plus some form of load
propagation.
> 
> Cheers,
> Michael

_______________________________________________
Tech mailing list
[email protected]
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/tech

signature.asc (application/pgp-signature, 189 B)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFFgv16A9rUluQ9pFARAkbhAKCDd1RlcsuA/N49kzsRjknQUh9JQgCdHvyM
Couf85VW2nTus/K3NnpKIyc=
=EpHW
-----END PGP SIGNATURE-----