[RC5] Filling the gaps

Rick Martin rick at housemartin.force9.co.uk
Tue Nov 24 23:39:49 EST 1998


[With apologies if this has been well-covered elsewhere, but I've just
spent half an hour scouring archives and couldn't find it.  Please
direct me to any discussion which might already have taken place on
this subject]


We're covering the keyspace in a nearly linear fashion.  That is, a
part of the whole keyspace is allocated to someone, then that part is
assumed to be processed.  The reality is, of course, that some parts
will be trashed by dying hard disks (he says, sheepishly), some will
be lost by "dead participants" (which discussion I have read).  Plus
there are blocks which get "stale" and don't come back for absolutely
ages.

My idea to fill the gaps is relatively straightforward:  Those of us
with permanent connections (as I have from one machine but not from
the others) could look to a proxy which maintains the keyblocks which
have, supposedly, already been allocated for processing.  Since these
machines are permanently on, the blocks would be returned within a
reasonable timeframe, say a day or two, without the usual "stale
block" problems, and we could probably achieve 100% coverage of at
least a handful of the key superblocks.

Workable?
    Rick

-- 
Any spelling or other typing mistakes included in the above are
deliberate and are provided solely for the enjoyment of our more
discerning readers.

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list