Re: [RRG] Tunnel fragmentation/reassembly for RRG map-and-encaps architectures

[Yakov Rekhter <yakov@juniper.net>]

Tony,

> On Jan 8, 2008, at 4:32 PM, Yakov Rekhter wrote:
> 
> >> Correct.  We've shown that the growth in the speed of normal DRAM is
> >> slower than the prefix growth rate.  Thus, simply the act of writing
> >> a list of all prefixes to memory (e.g., at initial convergence time)
> >> is taking longer each year.
> >
> > That, of course, assumes that we *have to* use DRAM. But do we
> > really have to ?
> 
> 
> No, it's not a requirement.  If your employer chooses to implement  
> BGP with a disk drive, that's fine by me.  ;-)

Since you propose this, I would rather let your employer to do this :-)
  
> You can also go in the opposite direction and go to non-commodity  
> memories, but then you go off of the commodity cost curves. Once you  
> do that, it's hard to continue to sustain our growth rate at constant  
> cost, and that's simply another way of failing to scale.

What is a non-commodity memory today need not be non-commodity
memory tomorrow.

Moreover, what really matters is not whether "you go off of the
commodity cost curves", but how far are you off.

Finally, while the argument about "constant cost" focuses on CAPEX,
let's not forget OPEX. The cost of operating the new system could
completely overshadow all the OPEX considerations.

Yakov.

--
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