On Sat, Feb 21, 2015 at 08:48:29PM +0100, Juliusz Chroboczek wrote:
> (Recall that multicast is 2Mbit/s at the phy.  13ms for a full-size frame,
> not counting the cost of collisions.)

Ok. But we also need a way to support fast router redundancy IMHO.

Cheers
    Toerless

> > Something as simple as RIP also makes it very easy to embed it as
> > a library into apps
> 
> No sensible platform allows unpriviledged code to manipulate the routing
> tables.  But if you manage to solve this particular problem, the stub
> implementation of Babel is 1000 lines of portable C.
> 
> > Lets just say with <= 5 sec failover time.
> 
> Babel has consistently been measured as having failover times of 2 to
> 5 hello intervals over IBSS, with no link-layer support, independent of
> update interval (due to triggered updates and reactive route requests).
> The protocol supports Hello intervals down to 10ms, and allows varying
> Hello intervals dynamically (a feature not used by the current
> implementation).
> 
> -- Juliusz

-- 
---
Toerless Eckert, [email protected]

_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to