Re: [RRG] Opportunistic Topological Aggregation in the RIB->FIB Calculation?

["William Herrin" <bill@herrin.us>]

On Mon, Jul 21, 2008 at 2:39 PM, Tony Li <tony.li@tony.li> wrote:
> However, those optimizations are not going to impact the RIB and propagate
> throughout the net.  [They also can end up being computationally intensive,
> sliding the problem around on the time/space tradeoff curve, not necessarily
> solving it...]
>
> Again, the problem is not to address just the FIB, the point is to deal with
> the overall scalability of the whole control plane.

Tony,

I disagree.

The RIB cost grows with the width of the system times the number of
entries. The system width is staying relatively static (and that's
expected to continue), so the growth cost is close to linear with the
number of entries. We already know that a $5k ($2k if you're on a
budget) Linux server running Quagga can keep up with 10 times the
entries and churn. With PC hardware's price/performance doubling every
3 years, we haven't long to wait before it can easily handle 100 times
the entries and churn that we have today.

The FIB cost, on the other hand, grows with the number of entries
times the packets per second which have to be routed. Both numbers are
rising quickly and both threaten to move faster. At the bottom end we
can currently route about a gigabit through a server with only
commodity hardware. Beyond that, we need a TCAM or parallel SRAM-based
radix trees, both of which are quite expensive in terms of
manufacturing, electrical draw and heat dissipation. Worse, the static
ram cells used by both are lagging behind PCs overall in terms of
their rate of improvement. We've had 1-meg CPU caches since The Sparc
II in the early '90's. We're only now getting 3-megs per core on
current Xeons. That's only doubling every decade.

All of this means one simple thing: We'll hit the wall on the FIB's
capability long before we hit the wall on the BGP RIB's capability.

Absent a RIB+FIB solution that the operators are enthusiastic about,
it's worth the effort to consider approaches that only reduce the size
of the FIB.

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