Advantages of dividing an Ethernet LAN with a bridge

1. Smaller collision domains — each segment becomes an independent collision domain, reducing collisions and improving overall performance.
2. Traffic filtering — the bridge only forwards a frame to the other segment if the destination actually resides there, keeping local traffic local and reducing unnecessary congestion.
3. Increased effective bandwidth — since segments operate more independently, total usable bandwidth across the LAN increases.
4. Fault isolation — problems on one segment (e.g. excessive collisions) don't necessarily degrade the whole network.
5. Extended reach — lets a LAN span more devices/greater distance than a single segment's physical limits allow.

Why no need for CSMA/CD on a full-duplex Ethernet LAN

CSMA/CD exists to handle collisions on a shared medium. In full-duplex Ethernet, each device has a dedicated point-to-point link to the switch with separate transmit/receive paths, so no other device contends for that same link — collisions cannot physically occur. Since the precondition for a collision doesn't exist, CSMA/CD's sensing and backoff mechanism becomes unnecessary and is disabled.

Why the frame with source address 43:78:6C:DE:10:00 was discarded

Convert the first byte to binary:

43 (hex) = 0100 0011 (binary)

In Ethernet addressing, the least significant bit of the first byte indicates whether the address is:

• 0 → a unicast (individual) address
• 1 → a multicast/group address

Here, the least significant bit is 1 (...0011, rightmost bit = 1), meaning this address is formatted as a multicast/group address.

However, a frame's source address must always be a unicast (individual) address — only a destination address is allowed to be multicast or broadcast; a frame can never legitimately originate from a "group." Since this address was presented as the source address but is bit-pattern-flagged as multicast, the receiver correctly recognizes the frame as malformed/invalid and discards it.