CHAPTER
4
Ethernet Basic
WHAT IS THE ETHERNET?
Ethernet:
A
family
of network technologies (standards) to build a LAN
Logical
bus
topology
Bus
physical
topology: The “older” Ethernet
Using coaxial cable as the bus cable
Star
physical
topology: The “newer” Ethernet (“
Star Bus
topology”)
Using UTP or Fiber Optic cable
OLDER VERSUS NEWER ETHERNET
Older: BUS Physical Topology
Newer: STAR Physical Topology
Older: Coaxial Cable as the Bus Cable
Newer: UTP Cable as the Patch
Cable and the Horizontal Run
WHAT IS THE ETHERNET?
Ethernet:
Ethernet frame
CSMA/CD
as the
network access method/media
access control
, i.e.,
a method to determine which
computer in the network can access the wire (in the
network) at a given moment
ETHERNET IN THE OSI 7 LAYER MODEL
ETHERNET FRAME
Ethernet frame
Created, sent, received, read, and destroyed by
Ethernet NIC
The advantages of using frame:
Frame defines the maximum data size, and therefore, large
data must be broken into smaller pieces. This provides two
benefits:
Benefit 1:
Preventing one computer from monopolizing
the shared wire in
the network (distributing the network use among all computers in the
network more evenly)
Benefit 2:
Whenever data is corrupted during transmission, the sending
computer only has to retransmit one (or few) frame(s) that contains the
corrupted data
ETHERNET FRAME
Seven major parts in Ethernet frame:
Part 1:
Preamble
64 bits of alternating 1s and 0s and ends with 11
Functions:
Enabling the receiving NIC to recognize the beginning of
a frame
Giving time to the receiving NIC to realize a coming of a
frame and therefore, can make all necessary
preparations to receive the frame properly
ETHERNET FRAME
Seven major parts in Ethernet frame:
Part 2 and 3:
Receiver/destination MAC address and
Sender MAC Address
A program/software called
SNIFFER
can enable NIC to run in
PROMISCUOUS MODE
to
process
ALL frames it receives,
regardless of their destination MAC addresses
http://www.tamos.com/htmlhelp/monitoring/monitoringusinghubs.htm
Hub
Computer A
Computer B
With “Packet Sniffer” this
computer will accept and
process the message
Accept the message
Drop/destroy the message
Drop/destroy the message
Drop/destroy the message
For Computer B
ETHERNET FRAME
Seven major parts in Ethernet frame:
Part 4:
Length
Information about the number of BYTES of DATA
inside the frame
Part 5:
Data
Data that is received from the upper layer of the OSI
seven
-
layer model (Network Layer): PACKET
Minimum number of bytes of data: 64 bytes
Maximum number of bytes of data: 1500 bytes
ETHERNET FRAME
Seven major parts in Ethernet frame:
Part 6:
Pad
If the data size is less than 64 bytes, the sender will
add more bytes in this part (PAD) to bring the data
up to the minimum 64 bytes
Part 7:
Frame Check Sequence
Error Detection Information (CRC for the Ethernet)
To detect data transmission error that causes data
modification during the transmission
CSMA/CD
Carrier Sense Multiple Access (CSMA)/Collision Detection
(CD)
Network access method (media access control)
Carrier Sense:
Every computer in the network must first examine (
sense
) the
BUS/SEGMENT before sending a frame
If the computer detects traffic in the BUS/SEGMENT, it will wait
for a moment, and then reexamine (re
sense
) the
BUS/SEGMENT
If the computer detects no traffic, it will send the frame to the
BUS/SEGMENT
CSMA/CD
Multiple Access:
All computers
(NICs) in a network perform CSMA/CD, hence,
have
equal access
to the wire (
no computer
has
greater access
(priority)
to the wire)
Collision Detection:
When
two or more computers
are sending frames to a network
simultaneously, a
collision
occurs, and all frames will be
corrupted/damaged/lost (the transmission is wasted)
During a frame transmission
(after sending a frame to a network),
the sender’s NIC will always monitor
(listen to) the network
If the sender’s NIC detects something else being transmitted in the
network, it recognizes that its frame has been corrupted by collision with
another frame
EARLY
ETHERNET
NETWORKS
10BaseT =
UTP cable (the “T”) based Ethernet
10BaseFL =
Fiber Optic cable (the “FL”) based
Ethernet
Using hub
10BASET: CABLING
CAT3 or higher
,
two pairs
,
UTP cable
RJ
-
45 connectors
Also known as CRIMP
Connecting UTP cable to RJ
-
45 connector: CRIMPING
Crimping standard (see Figure 4.12 on Page 67)
TIA/EIA 568 A
standard
TIA/EIA 568 B
standard
10BASET: SPECIFICATIONS
10
Mbps
Baseband
Maximum distance between the hub and the node
(computer/NIC):
100
meters
Maximum number of nodes per hub:
1024
Network topology:
Star Bus
Cable:
UTP, CAT
3
or higher, two pairs, RJ
-
45
connectors
10BASEFL/10BASEF: SPECIFICATIONS
10
Mbps
Baseband
Maximum distance between the hub and the node
(computer/NIC):
2000
meters
(
2
kilometers)
Network topology:
Star Bus
Cable: Multimode
Fiber Optic
10BASEFL/10BASEF: SPECIFICATIONS
Advantages:
Longer cable without repeater
Immune to electromagnetic interference (EMI)
More difficult to tap
More scalable
Scalability = ability to be adapt to (to be used with)
a new, faster technology
Disadvantages:
More expensive
More difficult to install
More fragile
MULTIPLE ETHERNET HUBS
Why using multiple Ethernet Hubs (multiple
segments)?
If one hub is down, the other hub(s) still works
→
NO SINGLE POINT OF FAILURE IN THE HUB
MULTIPLE ETHERNET HUBS
How to connect multiple Ethernet Hubs (multiple Ethernet
switches)?
Crossover cable
(UTP, two pairs, CAT
3
or higher): reversing the
sending and receiving pairs on the end of the cable (see page
70
)
One end of a cable is connected to the
regular port
, the other end is
connected to the
regular port
at the other hub
Crossover Port
(also called
Crossover, Uplink, In
-
port, Out
-
port
):
Use
regular cable
, NOT crossover cable
One end of the regular cable (
NOT crossover cable
) is connected to the
Crossover Port
, the other end is connected to the
regular port
at the
other hub
Autosensing Port
(a regular port that can turn to be Crossover
Port automatically whenever necessary)
One end of a regular cable is connected to the
Autosensing Port
, the
other end is connected to the
regular port
at the other hub
CONNECTING ETHERNET SEGMENTS
MULTIPLE ETHERNET HUBS
Hubs must be connected in
a single line (daisy
chain)
Why? This type of connection will simply extend the
“logical bus cable inside the hubs,” hence, maintain
the “logical bus topology”
Hubs
cannot
be connected in “
hierarchical
configuration
” (see page 70 figure 4.17)
Why? This type of connection will connect the “logical
bus cables inside the hubs” in a configuration other
than a “logical bus topology”
COLLISION DOMAIN
Collision Domain
(all computers are connected to
a
hub
(s)):
A network in which
there will be collision
if
two or more
computers
transmit frames simultaneously
Collisions can significantly slow down the network. Why?
Because every time there is a collision, the sender will
automatically
retransmit the frame (the original transmission
is wasted)
A
SWITCH
or
BRIDGE (i.e., a two
-
port switch)
can divide
a network into multiple collision domains
SWITCH
Computer A
Computer B
Accept the message
For Computer B
Switch
SWITCH
Switch does the switching by using its
lookup table
(i.e.,
an electronic database storing a list of ports and their
corresponding MAC addresses)
When a switch is just installed, as the lookup table is still
empty, it acts like a hub. However, every time it receives a
frame in a port, it will store the port number and the
sender MAC address of the frame in the lookup table
(see this animation:
http://www.cisco.com/image/gif/paws/10607/lan
-
switch
-
transparent.swf)
Port Number
MAC Address
1
2
3
Abc
Def
Ghi
SWITCH
Connecting multiple switches:
Use the procedures to connect multiple hubs
Spanning Tree Protocol
Enabling a switch to detect
bridge loop
(i.e., three or more
switches that are connected to one another forming a loop) and
to automatically shut down one port to break the loop
The problem with bridge loop
Broadcast storm
COLLISION DOMAIN
FAQ:
Can I break a network into multiple “collision
domains” by using a Router?
Yes, you can
So, why do I use Switch instead of a Router?
Switch is cheaper (although the price of Router
continues going down)
Switch is easier to install (e.g., no (not much)
configuration is needed)
HUB (REPEATER) VERSUS SWITCH (BRIDGE) VERSUS ROUTER
When do we use hub (repeater), switch (bridge), and
router?
Hub
Switch
Router
Router can do what a switch can do, but it is more expensive and
more difficult to install (e.g., requires more configuration).
Only router can connect two different networks.
Switch: To connect computers within a LAN
Router: To connect one LAN to another LAN
If you want to install packet sniffer to monitor incoming traffics to
your network, you can get hub
More sophisticated,
more expensive,
more difficult to install
Enter the password to open this PDF file:
File name:
-
File size:
-
Title:
-
Author:
-
Subject:
-
Keywords:
-
Creation Date:
-
Modification Date:
-
Creator:
-
PDF Producer:
-
PDF Version:
-
Page Count:
-
Preparing document for printing…
0%
Σχόλια 0
Συνδεθείτε για να κοινοποιήσετε σχόλιο