[Click] ad hoc routing protocols + performance

[John Bicket]

That probably won't be a problem, or if it is a problem it should be
pretty easy to fix, unless you have a bunch (like 10,000) nodes. 
Most protocols seperate route selection from data forwarding, and 
unless you do some kind of expensive route computation 
per packet, you can decide when to update your routing tables and
amortize the cost of route calculation over a bunch of data packets.

--john


Michael Voorhaen [michael.voorhaen at ua.ac.be] wrote:
>Hi,
>
>For a european project we are looking at several ways to implement the 
>routing protocol and click is one of the possibilities. We are currently 
>targeting OLSR as the candidate protocol. We are wondering if running 
>the entire protocol in kernel space would put a burden on the 
>forwarding. The fear of one of our partners is that the constant 
>(re)computations of the routing tables would interfere with the 
>forwarding since everything would be running in one process inside the 
>kernel.
>Does anybody now if this could be a real problem and if so are there any 
>ways inside click to circumvent this problem?
>
>Best Regards,
>Michael
>
>-- 
>========================================================
>Michael Voorhaen
>========================================================
>Phd Student
>Dept. of Mathematics and Computer Sciences
>PATS (Performance Analysis of Telecommunication Systems)
>University of Antwerp
>Middelheimlaan 1
>2020 Antwerpen, Belgium
>G1.11
>Phone: 03/2653905
>Mail: michael.voorhaen at ua.ac.be
>========================================================
>
>_______________________________________________
>click mailing list
>click at amsterdam.lcs.mit.edu
>https://amsterdam.lcs.mit.edu/mailman/listinfo/click