[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RRG] Hosts, DFZ, purity & incremental deployment



On Sun, Mar 23, 2008 at 12:33 AM, Stephen Sprunk <stephen@sprunk.org> wrote:
> Thus spake "Tony Li" <tony.li@tony.li>
>
> > | 1 - A solution which requires host changes in both communicating
>  > |     hosts is not incrementally deployable, since benefits only
>  > |     accrue to a tiny proportion of end-users (people who use hosts)
>  > |     due to the fact that initially, very few hosts have the
>  > |     upgrades.
>  >
>  > Hmmm... Well, you and I have very different semantics assigned to
>  > "incremental deployability".  I would consider anything that could be
>  > rolled out one host at a time without breaking anything as being the
>  > maximal amount of incremental deployability.
>
>  I think this definition is too weak.  What is the incentive to deploy
>  something when the only benefit it offers today is "doesn't break the status
>  quo"?  Not deploying also offers that benefit at a lower cost.
>
>  The minimum bar, to me, is that the first site receives benefits when it
>  pays the cost of deployment.  Those benefits may grow when more sites
>  deploy, but they cannot be zero in the initial state or you'll never achieve
>  critical mass, i.e. nobody wants to pay the costs because they don't get
>  benefits until _everyone else_ also pays the costs, so you get a classic
>  chicken-and-egg result just like IPv6 still suffers from.

Stephen,

I don't think you understand. After the major vendors have implemented
our new protocol to a standard of quality superior to BGP (before
they've been paid the first dollar I might add), we're going to
convince all of the tier-1 networks to deploy it. After all, they're
sensible people who know our new protocol is good and right. Then
they'll let all their customers know that they have to upgrade because
BGP is going away. The new protocol will have to be dual-stack for a
few weeks of course; we can't just have a flag day. That's why it has
to be able to run alongside BGP on the same equipment.

Really Stephen, how did you think we were going to deploy a new
routing protocol? Start with small edge systems and arrange things so
that they receive immediate benefits from the upgrade even though none
of their neighbors has upgraded yet? That's crazy talk!

Regards,
Bill Herrin


-- 
William D. Herrin herrin@dirtside.com bill@herrin.us
3005 Crane Dr. Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004

--
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