[c-nsp] Getting ready to pull the trigger: RSP720/SUP720

[Jon Lewis]

On Mon, 6 Apr 2009, Rick Ernst wrote:

> I'm planning on collapsing the border/core into a pair of
> 7600/Sup720-3BXLs, and it looks like they will be almost idle with this
> amount of load.

That really depends on the features you enable.  Try doing full netflow on 
a sup720 doing a few hundred mbit's of traffic, and they're suddenly not 
so mighty.

> The problem I am running into is spec'ing the aggregation layer.  Almost
> all of our traffic is ethernet now, and all the interfaces need
> bi-drectional rate-limiting/traffic-shaping/policing.  We have a variable
> bandwidth model and need to cap traffic at 1Mbs granularity. 1,5, and
> 10Mbs connections are common, and 20,50,100Mbs connections exist with a
> 200Mbs pipe in process.

We've been using 3550's for years for this, as they have the ability to 
police in both directions, per port, at whatever granularity you like. 
The 3560, which was supposed to be an improvement/replacement for the 3550 
lost this ability, which really shocked me when I configured my first one.
It can do per-port output shaping, but the granularity kind of blows. 
You're limited to 1/N * port rate, where N is an integer from 0 to 65535. 
This gives plenty (actually a huge waste of range) of granularity at the 
low end of bandwidth, but at the high end, you're limited to full rate, 
50%, 33%, 25%, 20%, etc.  If I'm wrong here, I'd love to hear it and be 
told how to limit a 100mbit port to say 40mbit/s.

----------------------------------------------------------------------
  Jon Lewis                   |  I route
  Senior Network Engineer     |  therefore you are
  Atlantic Net                |
_________ http://www.lewis.org/~jlewis/pgp for PGP public key_________