[MVMA] WVMN Network changes coming to AREDN Nightly Builds

[Chuck Gelm NC8Q]

On 2/17/25 10:54, Orv Beach via groups.io wrote:
> The AREDN team is introducing a new networking technology into the
> nightly builds with the ultimate long term goal of replacing OLSR.
>
> OLSR has many faults which AREDN has lived with for a long time. For the
> last couple of years we’ve been looking at alternatives and making
> incremental steps in the codebase to allow us to introduce something
> new. We can finally do that by adding Babel
> (https://www.irif.fr/~jch/software/babel/) to AREDN.
>
> Babel has a number of qualities which make it good for AREDN. First,
> it’s a loop free protocol so, regardless of how the network is changing,
> routing loops will never form in the network. Second, it’s primarily a
> reactive protocol which sends changes to neighbors when needed rather
> than broadcasting its state continually. Third, the protocol understands
> the difference between wired, wireless, and tunneled links –  the three
> link types AREDN utilizes. Fourth, it’s a layer-3 routing protocol,
> which integrates easily with how AREDN already operates. Fifth, it’s
> highly configurable which will allow an optimal setup for our use case.
> Finally, it’s simple.
>
> We considered a number of options, and another contender was
> B.A.T.M.A.N. Advanced. Unfortunately this protocol is not a good fit for
> AREDN as it primarily focuses on level-2 wireless networking. AREDN
> needs a protocol which can do more. We evaluated how we could use BATMAN
> and it wasn’t simple, efficient or pretty.
>
> If you’re interested in more comparisons between Babel and other options
> there are many good presentations on YouTube. This is a great primer on
> Babel itself: https://www.youtube.com/watch?v=1zMDLVln3XM
>
> Babel will roll out into the nightly builds alongside OLSR and the two
> will operate in parallel. If there is a Babel-only path between two
> nodes in the network, then Babel routes will be used to send traffic. If
> not, then OLSR routes will be used. This will let us deploy Babel and
> examine its performance in our networks without disturbing nodes which
> are not running Babel.
>
> Important: Be aware that older radios with limited memory – the ones
> using the tiny firmware images – will not be able to support both
> protocols and should be eventually replaced.
>
>  Evaluating success will depend on a few things, and the more feedback
> from the community the better. Our goals are to provide a more stable
> network, better routing decisions, and lower network overhead. Assuming
> success, the time scale to replace OLSR will be measured in years. Once
> OLSR is gone, any node not running Babel will disconnect, and we don’t
> want to leave our users behind so we are giving this as much time as
> needed for the transition to the newer protocol.
>
> The AREDN development team