[c-nsp] Downsides of combining P and PE functions into a single box

[Mark Tinka]

On Wednesday, October 19, 2011 11:35:45 PM Jared Mauch 
wrote:

> The real question:
> 
> Are you selling customer links that are near to or equal
> to the size of your core links(s).

Certainly not, or else what's the point :-)?

> Anyone doing 10GE edge or looking at 100GE for
> customer-facing handoffs can save significant amounts of
> money by doing P/PE.  While there are tradeoffs, not
> having the cumulative cost of a packet being A+B+C and
> perhaps can be localized to a single device has value. 
> I'm surprised that Rolland doesn't see this as an
> optimization as it would be something the Arbor
> equipment could help you optimize.

Well, we always try to have orders of magnitude more 
bandwidth in the Aggregation and Core then the Access, and 
I'm guessing a few on this list do too, even if you're 
handing off 10Gbps interconnects to customers.

But it's not untrue that commercially, there are savings to 
be made when you collapse the core.

> While some may see these cost savings as inelegant, the
> idea of a core will continue to come under these
> pressures.  Keep in mind the fraction of a chassis you
> must allocate for these edge <-> core links and core <->
> core links.  These have real world costs.

Luckily, equipment these days can be used either for the 
edge or the core (which is why P/PE designs are feasible in 
the first place).

That means I won't be relegated to a box that won't scale 
well (cost-wise) as my core grows. The ASR9000 and MX-series 
routers are excellent core routers, if you aren't keen on 
paying the premium that the CRS and T4000 routers command.

More over, the Aggregation boxes are developing denser 
10Gbps line cards, 40Gbps and 100Gbps line cards faster than 
traditional core boxes. Gotta love the in-house BU fights at 
vendor homes.

> There's a
> reason everyone didn't go out there and load-up on OC768
> hardware and just stuck with N*10G.  The finances don't
> work out.

STM-64/OC-192 is just about as sensible as it got. Anyone 
not doing Ethernet for 10Gbps or above these days is simply 
asking for it.

> The core has mostly gone ethernet for the major carriers
> that have high traffic loads.  The extra cost of the
> ethernet frame header, preambles, etc.. are worth the
> cost savings when compared to the lower SDH/SONET
> overhead at a higher per port cost.

Agree on that, and this has certainly made the core much 
more affordable.

Mark.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
URL: <https://puck.nether.net/pipermail/cisco-nsp/attachments/20111019/4add5859/attachment-0001.pgp>