[Babel-users] Propagation of unknown sub-TLV?

[Lorenzo Ghiro]

>
> Perhaps you can describe your extension, so we can think about it together?


Yes of course! Glad to see this interest even towards noobs like me :)

Well, the best way to start is by sharing a preprint version of a paper
accepted at Infocom2018 in which we* propose a distributed algorithm
for the computation of Centrality.
Full details are provided in the paper, anyhow I will sketchly present
the algorithm and the extension we imagine for Babel right now.

The algorithm let every node in a network compute its own value of "Load
Centrality"
and afterwards let nodes disseminate computed indexes so that each node,
at convergence, is aware of the centrality index of all nodes in the
network.

According to the definition of Load Centrality provided in the paper, nodes
have
to send/distribute "contributions" for the centrality score of their
neighbors and,
to do this, they need to add some information to classical Distance-Vectors.

If a usual distance-vector element is <d, m>, with d to indicate the
destination and m
a distance in a given metric, we want to also add <nh, loadOut, load>:
- nh is a list of potential next-hops, all at the same distance from the
advertised destination;
- loadOut is the Centrality contribution sent to neighbors by the sending
node;
- load is the Centrality Index of the destination, as far as the sender
knows.

With nodes/destinations we mean routers, indentified by router-ids, so we
think to put
the algorithm related information in a Router-id subtlv.
For the moment forget that actually Updates advertise distances associated
to prefixes,
not to router-ids...we just wanted to work on a network graph-layer.

So, only with little workarounds, we were able to run several emulations
with all nodes
running an extended version of Babel with provision for Centrality and, to
extend
our work, we were wondering how the Centrality computation would behave in
a network
with both standard and extended nodes. We guess very bad, in particular if
a "cutting-point"
node is not able to propagate contributions. So it would be nice, for our
purpose but maybe
also for others, if standard-nodes would support "transitive attributes".

Do you think the "transitive attributes" feature could be of general
interest? I strongly hope so :)

Best regards,

Lorenzo Ghiro

*Advanced Networking System group, DISI, University of Trento

2018-01-24 19:13 GMT+01:00 Juliusz Chroboczek <jch at irif.fr>:

> > I would like that Babel-nodes running an implementation of the original
> > protocol (call them standard-nodes) would be able to "forward
> > information" contained in sub-TLVs coming from "extended-nodes".
>
> [...]
>
> > Is it possible to accomplish this kind of “silent-propagation”?
>
> No.  In Babel, an unknown sub-TLV :
>
>   - is silently dropped if it is non-mandatory;
>   - causes the whole enclosing TLV to be dropped if it is mandatory.
>
> What you want is BGP's "transitive attributes".  There's no such feature
> in Babel.
>
> > For instance, I am thinking to this in order to deploy an extension
> > without the need to reboot the whole network.
>
> Perhaps you can describe your extension, so we can think about it together?
>
> -- Juliusz
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/babel-users/attachments/20180125/c56b29fe/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: preprint "On the Distributed Computation of Load Centrality?nd Its Application to DV Routing".pdf
Type: application/pdf
Size: 863312 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/babel-users/attachments/20180125/c56b29fe/attachment-0001.pdf>