[c-nsp] Cisco's new 4500-X 10G Aggregation Switches

[Skeeve Stevens]

So who is at fault here?  Cisco for not using bigger chips?

It sucks that we're being forced forward to IPv6, which is often requiring
large spend in new kit, but now that kit is going to perform at half the
throughput?

Seems crap to me.

*Skeeve Stevens, CEO*
eintellego Pty Ltd
skeeve at eintellego.net ; www.eintellego.net <http://www.eintellego.net.au>

Phone: 1300 753 383 ; Fax: (+612) 8572 9954

Cell +61 (0)414 753 383 ; skype://skeeve

facebook.com/eintellego

twitter.com/networkceoau ; www.linkedin.com/in/skeeve

PO Box 7726, Baulkham Hills, NSW 1755 Australia

The Experts Who The Experts Call
Juniper - Cisco – Brocade - IBM



On Thu, Feb 16, 2012 at 20:31, Lincoln Dale <ltd at aristanetworks.com> wrote:

> On 16/02/2012, at 5:06 PM, Saku Ytti <saku at ytti.fi> wrote:
>
> > On (2012-02-16 17:44 +1100), Skeeve Stevens wrote:
> >
> >> Just a question with the 4500-X.  Why would the routing performance be
> >> halved for v4 vs v6?
> >>
> >> IPv4 Routing Performance: 245Mpps
> >> IPv6 Routing Performance: 122Mpps
> >
> > I'd like to understand how could you technically make IPv4 and IPv6
> lookup
> > numbers same. Short of artificially slowing down or reporting IPv4 rates
> > conservatively.
> >
> > What type of mtrie stride could possibly do this? IPv4 8-8-8-8 and IPV6
> > 16-16-16-16-16-16-16-16, this would make IPv6 mtrie depth and width 2x of
> > IPV4.
>
> Stride doesn't come into it as its not likely a mtrie lookup but
> rather FIB is a TCAM.
>
> Performance is therefore less because to do a 144 bit lookup rather
> than 72 bit key (IPV4) is logically a double-wide lookup.
>
> Not all platforms have to do that but many do.
>
>
> cheers, lincoln.
>
> >
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>