Robert,
On 21/08/2021 19:32, Robert Raszuk wrote:
Les,
I beg to differ. ____
*/[LES:] Your statement suggests (and I am certain you do not mean
to do so) that if only we had the foresight to define the A-bit in
RFC 8919/8920 that we could have introduced support for Flex-Algo
without writing any new code at all. /**/đ/*
Nope. I meant to say that having support for flex-algo and FAD I can
think of new applications without need to write a new router's code
(assuming existing code is flexible enough :).
After all, aren't we in the era of network programming ?
*//**/____/*
*/We are talking here about a NEW APPLICATION. The only existing
example we have today (post RFC 8919/8920) is Flex-Algo â but given
time I expect some other application will come along. What that will
be and what kinds of calculations/attributes it might require is
something we do not currently know. The syntax used to advertise
link attributes does not magically allow new applications to be
supported without adding new code to the nodes./*
See above.
And I will perhaps say it again that to me flex-algo is more of a
mechanism to build new applications then NEW APPLICATION itself.
no, flex-algo is a single application, it's not a mechanism to create
new applications. The fact that you can create many constraints
topologies using flex-algo, does not mean these should be considered as
different apps. You have to put and keep clear borders at clear places.
We have them defined by ASLA and by base flex-algo draft.
thanks,
Peter
That is why if we would adhere well to ASLA architecture there should be
many more bits for it then one.
Thx,
R.
_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr
