How does cut-through switching work, and what is the Fragment Free variant?

Cut-through starts forwarding a frame as soon as it reads the destination MAC (Media Access Control) address (the first 6-byte field after the preamble), without waiting for the rest of the frame. Fragment Free waits for the first 64 bytes to filter out collision fragments (runts).

* Cut-through vs the Fragment Free variant. *

Standard Cut-Through:

• Reads only the destination MAC (the first 6-byte field of the frame, right after the preamble)
• Immediately looks up the egress port and begins forwarding
• The rest of the frame streams through while the beginning is already being sent out

Fragment Free (modified cut-through):

• Waits for the first 64 bytes before forwarding
• Why 64 bytes? Because Ethernet collisions can only corrupt frames shorter than 64 bytes (these are called runts)
• This filters out the most common type of error while still being faster than store-and-forward

Drawbacks of cut-through:

• No FCS (Frame Check Sequence) check → corrupt frames get forwarded, wasting bandwidth
• Cannot bridge speed mismatches → no buffering between different port speeds
• If error rates are high, cut-through can make congestion worse by propagating bad frames

Tip: Think of cut-through like forwarding a letter after just reading the address on the envelope — fast, but you have no idea if the contents are damaged.

Go deeper:

• Cut-through switching (Wikipedia) — forwarding before the full frame arrives, the store-and-forward contrast, and the 64-byte Fragment Free middle ground.