Thanks for all replies! So in my understanding the "8" is some number derived by Cisco Systems in their own unrevealed environment for which it guarantees catalysts proper functionality without any problems. Probably this environment can be imagined as L3 configuration with most features configured. The only limitation is TCAM capacity and theoretically we can jump above recommended number of SVIs to the prejudice of turning on some L3 features which utilize TCAM space. In reality the only way to know is to try :) Ok, thanks everybody again.
Alex. -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Peter Rathlev Sent: Monday, December 03, 2007 4:01 PM To: cisco-nsp Subject: Re: [c-nsp] Question about "show sdm prefer" command outputon Cat3560G On Mon, 2007-12-03 at 15:13 +0300, Alex A. Pavlenko wrote: > Dear colleagues! > > Consider simple output of "show sdm prefer" command on cat3560G > LOP3-1#sh sdm pref The current template is "desktop default" > template. > The selected template optimizes the resources in the switch to > support this level of features for > 8 routed interfaces and 1024 VLANs. > ...the rest is omitted > The question is what does "8 routed interfaces" exactly mean? As it says, with the level of features described (512 QoS ACEs and so on) it supports 8 layer 3 interfaces. If you do not use all the features, you can have more interfaces with no problems. > Is it a maximum number of SVIs or routed ports that I can configure on > a switch? There problably is, but the limit is quite large. I lab-tested creating 600 SVIs on a C3650G and it didn't complain. Since it does hardware forwarding you shouldn't feel any performance penalty from this. But if you start using features like ACLs, QoS, VRFs and so on, it probably can't handle this. > What happens when I configure 9 or more SVIs? If you configure too many interfaces / features, the switch will complain about lack of TCAM space. I haven't tried, but maybe it will start process-switching and you wouldn't want that performance wise, since it's designed as a hardware switch and the processer thus is relatively slow. Regards, Peter Rathlev _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
