Ethernet Switching

Question 1

What is the size of ethernet frames?

Answer 1

Minimum 64 bytes, maximum 1518 bytes

Question 2

What is an ethernet frame smaller than 64 bytes called?

Answer 2

Collision fragment or runt frame

Question 3

What is a frame more than 1500 bytes called?

Answer 3

Jumbo or baby giant frame

Question 4

What happens to ethernet frames that are too small or too big?

Answer 4

They are discarded

Question 5

Properties of a MAC address?

Answer 5

6 bytes in length
Must be unique
FF-FF-FF-FF-FF-FF

Question 6

What is the unique 6 hexadecimal code (3 bytes) at the start of a MAC address called?

Answer 6

Organisationally Unique Identifier (OUI)

Question 7

What is the last 3 bytes of a MAC address called?

Answer 7

Vendor code

Question 8

What happens if the MAC address destination doesn’t match the physical MAC address of the device?

Answer 8

The device discards the frame

Question 9

IPv4 and IPv6 name of protocol used to resolve IP addresses to MAC addresses?

Answer 9

Address Resolution Protocol (ARP) - IPv4
Neighbour Discovery (ND) - IPv6

Question 10

What are the 3 ways of sending a message?

Answer 10

Unicast - MAC address matches destination IP address
Broadcast - FF-FF-FF-FF-FF-FF - all hosts in local network receive packet
Multicast - 01-00-5E (IPv4) or 33-33 (IPv6) - hosts in the same multicast group will receive the packet

Question 11

How is a MAC address table built

Answer 11

By examining the source MAC addresses and port number of the frames received.

Question 12

How long is a MAC address table entry kept?

Answer 12

5 mins.

Question 13

Can multiple MAC addresses be associated with a single port?

Answer 13

Yes

Question 14

When will the MAC address be the default gateway

Answer 14

When the destination IP address is on a remote network

Question 15

What are the two different methods of switch forwarding?

Answer 15

Store-and-forward switching

Cut-through switching

Question 16

What is the name of the error detection algorithm used in store-and-forward switching?

Answer 16

CRC - cyclic redundancy check

Question 17

Describe store-and-forward switching

Answer 17

Receives entire frame and checks for errors before forwarding the frame. Discards frame if there is an error. Required for QoS analysis.

Question 18

Describe cut-through switching

Answer 18

Forwards frame before it is fully received, at least once destination address is read. No error checking performed.

Question 19

What are the two types of cut-through switching?

Answer 19

Fast-forward switching

Fragment-free switching

Question 20

Describe fast-forward switching

Answer 20

Lowest level of latency
Typical cut-through switching method
Packets may be relayed with errors, NIC will discard if so.

Question 21

Describe fragment-free switching

Answer 21

Stores first 64 bytes before forwarding. Most network errors and collisions occur during the first 64 bytes. Performs small error check on first 64 bytes before forwarding frame.
Compromise between store-and-forward switching and fast-forward switching

Question 22

What are the two types of memory buffering

Answer 22

Port-based memory - frames stored in queues at ports, possible for single frame to delay transmission of all frames due to busy destination port
Shared memory - deposits all frames into a common memory buffer shared by all switch ports. amount of buffer memory is dynamically allocated.

Question 23

What is autonegotiation?

Answer 23

Optional function which allows best speed and duplex options to be selected based on capabilities of the device.