Module 8

Ethernet Switching

Version 3

Ethernet Switching
Ethernet is a shared media
One node can transmit data at a time

More nodes increases the demands on the

available bandwidth
The probability of collisions increases, resulting in
more retransmissions

A solution to the problem is to segment .

Segmenting creates more collision domains
Version 3

Shared Media Environment

Shared media environment
multiple hosts have access to the same medium

Extended shared media environment

Using networking devices extends the environment
to accommodate multiple access or longer cable
distances

Point-to-point network environment

one device is connected to only one other device
(ex. dialup network connections)
Version 3

Shared media environments

Version 3

Layer 1 Devices
Layer 1 devices
repeaters and hubs

Extend collision domains

Primary function is extending cable segments
Additional hosts increase the amount of traffic
More traffic = greater chances of collisions
This results in diminished performance

Version 3

Repeater Rule
Four repeater rule:
No more than four repeaters between any
two computers
Contributing Factors
Repeater latency
Propagation delay
NIC latency
Late collision frames add delay that is
referred to as consumption delay
Version 3

Collision Domains

Collision Domains
Connected physical network segments
where collisions can occur

Collisions cause:
The network to be inefficient
Transmissions to stops for a period of time

Version 3

Collision domains

Version 3

Collision Domains
The types of devices that interconnect the
media segments define collision domains
Classified as OSI Layer 1, 2 or 3 devices
Layer 1 devices do not break up collision
domains
Layer 2 and Layer 3 devices break up
collision domains
Increasing the number of collision domains is
known as segmentation

Version 3

Segmentation

Version 3

10

Network segment

Version 3

11

Layer 2 Devices
Layer 2 devices
Bridges and Switches
Segments collision domains
Controls frame propagation using the MAC
address
Tracks the MAC addresses and segment they are
on
Version 3

12

Layer 2 Bridging

Version 3

13

Bridges
Has only two ports and divides a collision domain
into two parts
Entire network will share the same logical
broadcast address space
Creates more collision domains but will not add
broadcast domains
All decisions made are based on MAC or Layer 2
addressing
No effect on the logical or Layer 3 addressing

Version 3

14

Layer 2 Switching

Version 3

15

Switches
A switch is a fast, multi-port bridge
Each port creates its own collision domain
A switch dynamically builds and maintains a ContentAddressable Memory (CAM) table
The CAM holds all of the necessary MAC information
for each port

Version 3

16

Switch Operation
Micro-segments consist of the switch port and the
host connected to it
Communication in both directions at once is known
as full duplex
Most switches are capable of supporting full duplex,
as are most network interface cards (NICs)

In full duplex mode, there is no contention for

the media.
A collision domain no longer exists
Theoretically, the bandwidth is doubled when
using full duplex
Version 3

17

Version 3

18

Switch Modes
Cut-through switching
A switch transfers the frame as soon as the
destination MAC address is received
lowest latency
no error checking

Version 3

19

Switch Modes
Store-and-forward switching
Higher latency
The switch receives the entire frame before
sending it out
Verifies the Frame Check Sum (FCS)
Invalid frames are discarded at the switch
Version 3

20

Switch Modes
Fragment-free switching
A compromise between cut-through and
store-and-forward switching
Switching begins before the entire data field
and checksum are read

Reads the first 64 bytes

Including the frame header
Verifies the reliability of:
Addressing
Logical Link Control (LLC) protocol
Version 3

21

Switch Modes
Synchronous switching
The source port and destination port must

be operating at the same bit rate

Asynchronous switching
The bit rates are not the same
The frame must be stored at one bit rate
before it is sent out at the other bit rate
Store-and-forward must be used
Version 3

22

Switch Modes
Asymmetric switching
Switched connections between ports of unlike
bandwidths
Asymmetric switching is optimized for client/server
A server requires more bandwidth dedicated to the
server port to prevent a bottleneck at that port

Version 3

23

Spanning Tree Protocol

Switching loops can lead to broadcast storms that will
overwhelm a network.
To counteract loops, switches are provided with the
Spanning-Tree Protocol (STP)
Switches in a LAN using STP
Send Bridge Protocol Data Units (BPDUs) out all its
ports
Lets other switches know of its existence
Elect a root bridge (switch) for the network
Switches use the Spanning-Tree Algorithm (STA) to
resolve and shut down the redundant paths
Version 3

24

STP
Each port using Spanning-Tree Protocol is in
one of the following five states:
Blocking
Listening
Learning
Forwarding
Disabled

Version 3

25

STP
A port moves through five states as follows:
From initialization to blocking
From blocking to listening or to disabled
From listening to learning or to disabled
From learning to forwarding or to disabled
From forwarding to disabled

Resolving and eliminating loops creates a logical

hierarchical tree with no loops
The alternate paths are available if needed
Version 3

26

Spanning tree protocol

Version 3

27

Layer 2 Broadcasts
Ethernet Broadcasts
When a node needs to communicate with all
hosts on the network
A broadcast frame with a destination MAC
address 0xFFFFFFFFFFFF is sent
The network interface card (NIC) of every
host must respond
Version 3

28

Layer 2 Broadcasts
Layer 2 devices must flood all broadcast and
multicast traffic
Broadcast Radiation
The accumulation of broadcast and multicast
traffic from each device

Broadcast storm
Circulation of broadcast radiation that saturates
the network
There is no bandwidth left for application data

Version 3

29

Layer 2 Broadcasts
The three sources of broadcasts and
multicasts:
Workstations
Routers
Multicast Applications

Version 3

30

Broadcast & Collision Domain

Collision Domain

Collision Domain

Version 3

31

Layer 3 Devices
Layer 3 devices
Routers
Do not forward collisions
Breaks up collision domains
Broadcast domains are controlled

Version 3

32

Broadcast domain

Version 3

33

Broadcast Domain
Broadcast Domain
A grouping of collision domains
All the nodes that are a part of that network segment
bounded by a layer three device
Broadcasts have to be controlled at Layer 3 devices
Layer 2 and Layer 1 devices do not control
broadcasts

Version 3

34

Data Flow
Layer 2 devices filter data frames based on the
destination MAC address
A Layer 2 device will forward the frame unless something
prevents it from doing so

Layer 3 devices filter data packets based on IP

destination address
A Layer 3 device will not forward the frame unless it has to
Layer 3 device creates multiple collision and broadcast
domains

Version 3

35

Dataflow

Version 3

36

Latency
The delay between the time a frame leaves the source device
and the time the frame reaches its destination

The following conditions can cause delays:

Physical media
Circuit delays
Electronics that process the signal along the path
Software delays
Decisions that must be made to implement switching and
protocols
Delays caused by the content of the frame
Destination MAC address has to be read

Version 3

37

Latency

Version 3

38