North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: layer 3 switch debate

  • From: Richard A Steenbergen
  • Date: Thu Sep 26 16:36:35 2002

On Thu, Sep 26, 2002 at 11:38:38AM -0700, ip dude wrote:
> 
> Does this rule of thumb still apply considering the modern layer 3
> switches available? If not, why? What makes a layer 3 switch
> sub-standard to a pure router? Any quantitative analysis you could
> provide would be greatly appreciated.

Layer 3 is layer 3, whether you use a general purpose processor, a
specialized asic, or a cam. A "layer 3 switch" is really just a product
from a traditional switch vendor who wants to get into the "layer 3"
market.

They all forward millions of packets per second when the cams are
programmed and things are simple, that means nothing. What sets a good
vendor apart from a substandard vendor has very little to do with how well
it forwards frames or packets in the lab. It is how they implement every
routing protocol, cli, management functionality etc, CORRECTLY and
reliably. It is how they handle real-life or exceptional conditions, like
random-dst traffic which stress the initial route lookup and cam
programming operations.

There are a lot of companies who want to make core routers or layer 3
switches or whatever marketing calls them, but they almost all fail when
it comes to implementing those pesky little things like routing protocols.
In my opinion there are only two vendors who meet the test right now,
Cisco and Juniper. If you think anyone else can correctly function as a
core router, you don't have a real core, and all the koolaid in the world
won't change that. :)

-- 
Richard A Steenbergen <[email protected]>       http://www.e-gerbil.net/ras
PGP Key ID: 0x138EA177  (67 29 D7 BC E8 18 3E DA  B2 46 B3 D8 14 36 FE B6)