Question about Campus Switching Design

[Neuromancer]

I'm studying for BCMSN and have read in several places this statement:

"Devices in a campus network's core layer or backbone should be optimized for high-performance Layer 2 or Layer 3 switching."

Why would you go from layer 3 in the distro layer back to layer 2 in the core? Does anybody acutally use this methodology in their networks? Is layer 3 processing considered "expensive" compared to layer 2?

Thanks!

Chris

[Edison Ortiz]

I believe what they are trying to convey is that the core shouldn't be used for making routing nor filtering decisions (that's what the distribution switches are for).

The Core should be Layer3 capable but its main function is to forward the traffic (read: Layer3 switching above).

Just think of the Core as a typical FireWall. It's there to inspect the packet but you wouldn't have a FW as the default gateway making routing decisions for your network, would you ?