On 8/19/08 4:08 PM, Tony Li allegedly wrote: >> As to the third, a distributed mapping functionality is >> fundamentally a whole different - and easier - kettle of fish from >> a distributed path-selection computation. The latter is >> (necessarily) a computation, and depending on how it's split up for >> distribution, it can become quite painful. (The Destination Vector >> approach, where the computations of the algorithm are themselves is >> distributed, is the worst case.) Even the best cases are still far >> more complex than the distribution of a mapping, though. > > So then why are the ALT folks reusing distributed path-computation > for distributing a mapping?
(Adding in your correction that this is about an index to mappings, not the mappings themselves ...) - This is a virtual topology, so the topology exactly follows addressing, and information is highly aggregated. Churn is quickly damped. The computation is immensely simplified. - It allows mapping information to be distributed from sources that are directly connected with operational reality, at a time when that reality is happening -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
