On Thu, Nov 20, 2008 at 07:50:54PM +0100, <Matteo>:
~> Don't you think that all this ETP flying around the net, could overload the
~> network?
~> If all the nodes start an ETP in the same time, there could be some problem
~> right?
As you can read in the previous email (DART, Netsukuku) you shouldn't worry
about this problem too much (see Distance Vector routing).
However, let's do a simple counting.
Worst case: you have n nodes that join at the same time.
Then each node creates an ETP. In total there are n ETPs flying around.
So if you have 256 nodes, you'll get 256 different ETP.
Now, since the topology is hierarchical, you must count in each level.
So, suppose you are a node.
In the level 1 there 256 nodes (including you). So here you see 256 ETPs flying
around.
In the level 2 there 256 gnodes (including your). The ETP treats each gnode as
a single node.
So, even in level 2 you see only 256 different ETPs.
The same for the other levels.
So, if you have 4 levels, then
256 + 256 + 256 + 256 = 1024
different ETPs are created and fly around in the network.
NOTE: this isn't the number of ETPs that you receive. It is the number of
_different_ ETPs that are spawned/created/generated.
To calculate the number of received ETPs you must take into account the
actual configuration of the net.
~> ps.: I work with a kind of network that suffer this kind of problem, so, how
~> you solve that?
Remember: no mobile in Netsukuku (unless you find a good solution. if you want
I can give you some starting points).
Note again: this isn't a rigorous counting. It is raw, just to give you an
idea. For a rigorous counting you have to take into account a lot
of other elements, and I don't have time for that.
Last note: the real problem in Netsukuku is represented by how you assign the
IPs, not how you tell the nodes about the available routes.
Later.
--
:wq!
"I don't know nothing" The One Who reached the Thinking Matter '.'
[ Alpt --- Freaknet Medialab ]
[ GPG Key ID 441CF0EE ]
[ Key fingerprint = 8B02 26E8 831A 7BB9 81A9 5277 BFF8 037E 441C F0EE ]
_______________________________________________
Netsukuku mailing list
[email protected]
http://lists.dyne.org/mailman/listinfo/netsukuku
