From: Howard C. Berkowitz (hcb@gettcomm.com)
Date: Fri Apr 02 2004 - 11:28:27 GMT-3
At 4:23 PM -0500 3/31/04, Vazman@aol.com wrote:
>Just saw this update today on Cisco's site.
>
>http://www.cisco.com/warp/public/625/ccie/announcements/#secupdatespix
>
>- Vazman
>
It's always interesting to read between the lines of lab
announcements and try to guess what might be changing. From the above
announcement,
>On July 7, 2003 the CCIE program will be adding the 3700 series
>router to all CCIE lab locations worldwide. Please refer to the
>equipment list for detailed hardware specifications for each CCIE
>track.
Following this link, we find "3700 routers", which still could be
interpreted either as 3770 series L3 switches, or 37x5 series
multifunction devices.
What might be the lab implications? Let me speculate, emphatically
saying I have no knowledge of what Cisco actually plans.
Obviously, if these are added as additional devices, the potential
configurations become more complex. Historically, there's been a
tendency to replace existing devices. Let me first speak of the
possible implications of the devices themselves.
If they mean the 3770 L3 switch, there probably isn't much difference
in the device itself, which in many respects is a repackaging of the
3550, optimized for more dense voice applications.
If they mean the 3725 or 3745, things get more complicated. These can
contain switching modules, which would allow more complex switching
configurations (see below). These devices also can take AIM modules
that might need configuring. There are some new interface modules,
some of which are nothing new (e.g., L2 Ethernet), but some useful
oddballs like contact closure. I have real-world applications for
contact closure in monitoring medical equipment, but I can't see an
easy way to test these in the lab.
The big question is whether there will now be more than 2 devices
with full swiching capability. With two switches, a certain number
of advanced interswitch features can be expected, such as port
channels. For that matter, things like port channels could be run to
routers.
Three switches introduce the possibility of several additional
failover mechanisms. These mechanisms exist both as Cisco
proprietary and newer IEEE functional equivalents. The failover
mechanisms include (just using Cisco names) UplinkFast, BackboneFast,
RootGuard and BPDU Guard.
Beyond failover mechanisms, it might well be possible to have
considerably more complex VLAN scenarios, involving such things as
per-VLAN or per-VLAN-group spanning tree, VTP pruning, VTP
transparent and nontransparent modes, etc.
Is there good news about this? Probably, for home labs. You don't
necessarily need a 37xx to test these features (let's assume a
3-switch configuration). Even in the new IOS switches, a 2950 can
easily exercise most (but not all functions) if you have a 3550 or
two.
In many cases, you might even do well with a 5000/5500 or the like,
at least to test the Cisco version of some of these features -- you
will have to check feature support; there's at least some. You might
not be able to do the IEEE versions.
Thoughts?
This archive was generated by hypermail 2.1.4 : Mon May 03 2004 - 19:48:41 GMT-3