[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RRG] Tunnel fragmentation/reassembly for RRG map-and-encaps architectures
Yakov,
[Control plane v. Data plane]
Ok, let's gloss over details... After all, it is precisely
in the details where the devil is...
The relation to control plane and data plane isn't some fixed thing.
Even within the data plane you make decisions based on the data that is
in the packet. What you cache and where it is kept cannot itself
dictate a solution, because that leads to circular logic. We're here
because of precisely what we cache and where it is kept, and we need to
make a change. Anyone who thinks differently should be satisfied today
and for the future with what we have because that's all we'll have.
LISP offers a slightly different view of what needs to be managed
where. It takes very little imagination to understand how to manage the
new "control plane" elements in the data plane at line rate.
Consider a multi-homed site connected to two ISPs, ISP1 and ISP2.
The ETRs are maintained by ISPs. How would you get the information
about ETR failure via IGP in this scenario ?
The obvious answer is "some other way". There is an architectural
assumption that if you want fast failover between two ETRs they must
somehow communicate with each other. However, this is not the case I
would optimize first. I would expect the enterprise to manage both
ETRs, as the configuration complexity of doing so is looking to be
considerably lower than that of BGP.
In contrast
to this, with the current routing restoring reachability may require
(much) narrow scope of the the failure information propagation.
This has always been an option, but it's not solely used by any serious
enterprise.
You are mistaken - this is exactly how the Internet routing system
works today. Specifically, the time to restore reachability is
less then the time it takes to propagate the failure information
Internet-wide.
I disagree and I believe it is you who are mistaken. Enterprises want
more than just line redundancy. They want protection against a provider
failure and path diversity, as happens occasionally due to fiber cuts,
data center failures, and the like. The consumer space is even more
clear. Why should one install more lines into the home and pay
additional costs relating to those lines when the wires are already
there today, but happen to be separated by two carriers?
So why do you think it will be acceptable in the future?
Because it works.
I personally have an aversion to rejected technical solutions that are
known not to satisfy clear business cases.
Eliot
--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg