[j-nsp] TCAM full on EX8200?

[Richard A Steenbergen]

On Sat, Oct 22, 2011 at 08:19:33PM +0400, Pavel Lunin wrote:
> 
> As far as I understand, it's not really correct to compare difficulty 
> of these two operations, since they are performed by two different 
> units inside the chip. While label lookup instead of full IP table can 
> dramatically simplify the lookup unit's life, the unit, which inspects 
> packets and extracts bits from them, must be still quite complex even 
> for label-only router. Hashing ALU's life is not a peace of cake 
> either. Say, EX series PFE use only 6-bit seeds to construct hash on 
> them. In case you want to push a whole 20-bit label to the hash seed, 
> I'm afraid, you'll need more bits in ALU registers, more cycles or 
> something else.

Well, nobody said it was EASY, just EASIER. :)

The point is, just by going from full tables to not-full-tables you can 
start taking advantage of existing commodity chips which CAN do IP and 
some pretty deep hashing already, and move from thousands of dollars per 
10GE and 80-120G/slot (MX/ASR-style) down into hundreds of dollars per 
10GE and 480-640G/slot.

Personally my take is that PTX missed the mark as far as interesting 
target customer size goes. There are still a *LOT* of places where you 
could very easily replace a $1mil core T/CRS box at $retardecarrier with 
a $10-20k 1-2U box that does 64x10GE, or 40GE, etc. But obviously 
Juniper doesn't want to canibalize from their own T (or at worst MX) 
market, which is why only made the PTX product that they did (pick 
between massive or super-$%^&ing massive, and at only ~MX-style port 
prices).

Eventually someone will come along and realize that there is an untapped 
market here, and then the promise of cheap label switching routers will 
be fulfilled. IMHO the reasons this hasn't happened yet are a) only 
Cisco/Juniper *REALLY* have a good handle on the software side of MPLS, 
and b) not enough non-carriers are currently running (or even currently 
able to conceptualize running) label-only cores. Of course the large 
carriers with the most to gain also tend to be the least innovative and 
the most unwilling to consider cost when designing their architecture, 
until they go bankrupt of course. :)

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)