Destination MAC first and switch learning process

daan5000

Hello community

I have 2 questions regarding to how switches process ethernet frames:

First question: why does the destination MAC address come first in an Ethernet frame, while in an IP packet the source IP comes first? I once heard that the logic behind this is so that the switch can make its forwarding decision very early when it receives the frame, but that doesn't sound correct as it also needs to check the source MAC address for learning purposes.

Second question: what does a switch do first, does it first check the destination MAC address to make its forwarding decision, or does it first check the source MAC address to see whether or not it has previously learnt that address?

Thank you

Kind regards

Daan

clarson

Yes the ethernet frame had the destination mac address first so the switch can make its forwarding decision as quick as possible (cut-through switching). a switch doesn't need to know the source address to send the packet. just needs to know which port to send the packet. which would be a cam table lookup. flood if not in the cam table. just don't send out the same port it was received.

now days there are lots of security features that check the source address to see if the packet should be allowed on the network at all.