[Battlemesh] Battlemesh v5 tests

Clauz clauz at ninux.org
Mon Mar 12 15:56:00 CET 2012


On 03/11/2012 12:37 PM, Antonio Quartulli wrote:
> Hi Clauz,
> On Sun, Mar 11, 2012 at 12:03:08PM +0000, Clauz wrote:
>> On 03/11/2012 10:59 AM, Antonio Quartulli wrote:
>> Hi, Antonio.
> I didn't talk about moving NODES, but I talked about moving CLIENTS.
> E.g. in your ninux network could be a normal user which, for some reason, moves
> from an Access Point to the next one (imagine two close houses with two
> different ninux APs).

AFAIK the other protocols (or at least their open source
implementations) don't have (yet) this feature, so I don't think is the
case to include it in the "official" tests.

>>  * it might be not easy to repeat the test (i.e. walk many times the
>> very same path at the very same speed) for each protocol
> Well, since the idea is not to measure how well routes react (remember, we are not
> moving nodes), the test could simply consist in a deterministic number of
> roaming.
> E.g.: assuming to have a full mesh topology made up by 4 nodes, we can
> trigger a roaming event and let a client "move" from node A to node B and then
> check if connections are still alive. In case of dead connections, what types
> those are (e.g. extremely interactive protocols reacted bad while an FTP
> download didn't notice the roaming at all).
> This is just a roughly idea, but we can elaborate something more complex.

If we have the hardware, and if you want to take care of this, I don't
think that anybody will try to stop you from performing the tests and
giving a talk about the results :)


More information about the Battlemesh mailing list