[j-nsp] TCAM full on EX8200?

Phil Mayers p.mayers at imperial.ac.uk
Sat Oct 15 06:04:37 EDT 2011


On 10/14/2011 07:04 PM, Jeff Wheeler wrote:
> On Fri, Oct 14, 2011 at 3:52 AM, Michele Bergonzoni<bergonz at labs.it>  wrote:
>> can only be done with TCAM. For those who want more info on this issue, this
>> is the very interesting reference that I received in a private email:
>>
>> http://www.firstpr.com.au/ip/sram-ip-forwarding/
>
> I wouldn't use that particular document as a "reference."

"Take 1st 24 bits of the IP, use as an index into RAM, profit. IPv6 and 
longer than /24 left as an exercise for the reader"

Wow, that easy eh? Problem solved!

;o)

> When your data set grows, like the DFZ, it is less expensive to have
> an ordinary RAM and search algorithm.  Which particular kind of RAM is
> in use is not all that exciting, but in routers that use some kind of
> tree search, you can find all manner of SRAM and DRAM.

Right. So, because traditional destination-based IP routing lookup is 
generally fixed-length (i.e. at most 128 bits for IPv6) then performance 
of a RAM-based forwarding algorithm is relatively predictable.


>
> There has been some mention about TCAM being good at ACL matching.  It
<snip>
> The reason vendors do not want to do this is the TCAM can evaluate a
> very large ACL in one step, while a ALU+RAM might be more
> power-efficient, if you have a ACL with hundreds of entries, it will
> take a long time to process packets, and you will not have wire speed

...whereas because ACLs are variable length, determined by customers and 
possibly large, performance of a RAM-based ACL algorithm is hard to 
predict, and people want predictable performance, and usually line-rate 
performance.

Having said that - personally I might be willing to trade line-rate 
performance for (say) an ACL mechanism with near line-rate for simple 
ACLs, the option of a "jump" opcode, and some way of knowing what the 
exact performance (range) of a given ACL/interface combo would be.

Do I take it that non-destination-based routing (policy routing, filter 
based forwarding) are therefore implemented differently on boxes that 
use RAM-based forwarding?

> It is nice to understand how your routers work at a deeper level.

Indeed. I'm fairly familiar with our current generation of TCAM-based 
Cisco kit, but we've found less room for "real routers" in the network 
because of cost/performance considerations, so it's interesting to hear 
that EX8200 switches are RAM-based.

> More often than not, though, all it will do is make you wonder how a
> given product ever shipped or make you angry that you can't just get
> an MPLS P box for the same price as an Ethernet switch.  :-)

Hehe. "Tag switching will make core routers really cheap, you'll have a 
few really big PE routers only". Wasn't that the line we were sold with TDP?

It's particularly interesting to hear that RAM-based forwarding has 
potentially far lower power usage; presumably this is why e.g. Cisco are 
crowing about how "green" the ASR9k is. Higher UPS runtimes a side benefit!

Thanks for the informative replies everyone.

Cheers,
Phil


More information about the juniper-nsp mailing list