On Mon, 18 Feb 2002, Jeff Darcy wrote:

> > You don't consider having to route data though other nodes all the time a
> > slowdown?
> 
> There might be a latency cost for the *first node* to request data along a
> path, but that's far outweighed by the latency savings when *every
> subsequent requester* can get it from nearby rather than far away.  If you
> increase the first node's latency by 20%, and then save 50% for each of the
> next nine, your overall average latency drops by more than 40%.  This effect
> persists even for fairly modest cache hit rates.
> 
> There might be a lot of ways that you'd want to change away from the way
> Freenet has done things, but the caching strategy is probably not one of
> them.  A very large number of very intelligent people have done a lot of
> work trying to find better methods, but you'd be hard pressed to find any
> for which there's an efficiency gain to balance the additional complexity.

It is better to route data though other nodes even when there is a more 
direct path between two nodes?  In particular do you think that this 
strategy is flawed:

  I also believe that by simply transferring data from one node to another
  a lot of the benefits of freenet's caching algorithm can be retained.  
  Because, once node A transfers a key from node B nodes close to A (say
  B,C,D) can transfer data from node A instead of having to download it 
  from node S which is farther away.  If node A is inaccessible to node 
  B,C,D for some reason this scheme will still work because once node B 
  downloads the data nodes C and D can transfer it from node B.  The more 
  nodes that download the data the less chance there is that *all* off 
  them will be inaccessible to other nodes.

  However, if for some reason nodes A,B,C,D and completely inaccessible to 
  other nodes, node S will notice that it is getting a large number of 
  requested from far away and will upload the data to some other nodes 
  which are closer to the origin of the requests.

--- 
http://kevin.atkinson.dhs.org



_______________________________________________
freenet-tech mailing list
[EMAIL PROTECTED]
http://lists.freenetproject.org/mailman/listinfo/tech

Reply via email to