Transcript Bit oriented Protocols - Al al

Chapter 12
Local Area Networks
 A LAN is a data communication system that
allows a number of independent devices to
communicate directly with each other in a limited
geographic area.
 LANs are dominated by four architectures:
 Ethernet
 Token
Bus
 Token Ring
 FDDI
Standards of IEEE and part of
Its project 802
ANSI standard
1
Project 802
 It is a way of specifying functions of the physical
layer, the data link layer, and the network layer to
allow for
 interconnectivity
and compatibility of different LANs
 Data to be exchanged across incompatible networks
2
Project 802
 It has subdivided the data link layer into 2 sub layers
Logical Link Control LLC
 Medium Access Control MAC
 LLC is non-architecture-specific, that is, it is the same for all IEEE–
defined LANs
 The MAC sub layer, contains a number of distinct modules; each
carries proprietary information specific to the LAN product being
used

3
IEEE Project 802 model
takes the structure of HDLC frame and divides into two sets of functions
IEEE 802.2 LLC protocol
 It contains the end-user
portions of the frame:
 Logical address
 Control information
 Data
MAC
 It resolves the contention for
the shared media.
 It contains the following
specifications necessary to
move information:
 Physical address
 Synchronization
 Flag
 Flow control
 Error control
4
Protocol Data Unit PDU
 The data unit in the LLC level is called PDU.
 It contains four fields familiar to HDLC
DSAP used to identify the protocol stacks on R/S stations
 SSAP
 Control field
 Information field
 The PDU has no flag fields, no CRC, and no station
address, these fields are added in the MAC layer

5
PDU Control Field
6
Ethernet (IEEE 802.3)
 It defines two categories
 Baseband
- Specifies a digital signal
 Broadband - Specifies an analog signal
7
Access Method CSMA/CD
 Whenever multiple users have unregulated access to
a single line, there is a danger of signals overlapping
and destroying each other.
 Such overlaps, which turn the signals into unusable
noise, are called collisions.
 A LAN therefore needs a mechanism to minimize the
number of collisions, and maximize the number of
frames that are delivered successfully.
 This mechanism is called carrier sense multiple
access with collision detection (CSMA/CD).
8
Evolution of CSMA/CD
9
MAC Frame
10
Ethernet Segments
11
10BASE2
12
10BASET
13
Figure 12-13
14
Figure 12-13-continued
1BASE5
15
Figure 12-14
WCB/McGraw-Hill
16
Token Ring (802.5)
 Token ring resolves the problem of multiple
stations tries to capture the link at the same time.
 Each station may transmit only during its turn and
may send one frame per turn
17
Access Method
 The mechanism that coordinates this turn is called






Token Passing
Whenever the network is unoccupied, it circulates a 3byte token.
This token is passed from NIC to NIC in sequence
until it encounters a station with data to send.
The station waits for the token to enter its network
board.
If the token is free, the station may then send a data
frame.
It keeps the token and sets a bit inside its NIC as a
reminder that it has done so, then it sends its one data
frame.
The data frame proceeds around the ring, being
regenerated by each station
18
 Each intermediate station examines the destination




address, finds that the frame is addressed to another
station, and relays it to its neighbor.
The intended recipient recognizes its own address, copies
the message, checks for errors, and changes 4 bits in the
last byte of the frame to indicate address recognized and
frame copied.
The full packet then continues around the ring until it
returns to the station that sent it.
The sender receives the frame and recognizes itself in the
source address field. It then examines the addressrecognized bits. If they are set, it knows the frame was
received.
The sender then discards the used data frame and releases
the token back to the ring.
19
Figure 12-15
Token Passing
20
Figure 12-15-continued
Token Passing
21
Figure 12-15-continued
Token Passing
22
Figure 12-15-continued
Token Passing
23
Figure 12-16
Token Ring Frame
24
Priority and Reservation



Once the token has been released, the next station on
the ring with data to send has the right to take charge of
the ring.
In IEEE 802.5 model, it is possible that a busy token can
be reserved by a station waiting to transmit, regardless of
the station location on the ring.
Each station has priority code. As a frame passed by a
station waiting to transmit may reserve the next open
token by entering its priority code in the Access control
field (AC) of the token or the data frame


A station with a higher priority may remove a lower
priority reservation and replace its own.
If station has equal priority, the process is 1st–come, 1st–
serve
25
Monitor Stations







One station on the ring is designated as a monitor station.
The monitor sets a timer each time the token passes.
If the token does not appear in the allotted time, it is
presumed to be lost and the monitor generates a new
token and introduce it to the ring.
The monitor guards against recirculation data frames by
setting a bit in AC field of each frame.
As a frame passes, the monitor checks the status field.
If the status bit is set, it knows that the packet has
already been around the ring and should be discarded.
If the monitor fails, a second station, designated as back
up, takes over
26
Token Frame
Is a placeholder and reservation frame.
 SD: indicates frame is coming
 AC: indicates that frame is a Token and
includes the priority & reservation fields.
 ED: indicates the end of the frame
27
Abort Frame
 Generated by the sender to stop its own
transmission or by the monitor to purge an old
transmission.
 Token and Abort frames are both truncated
data/command frames.
28
Figure 12-18
Token Ring
29
Implementation of the Ring:
 The ring in a Token ring consists of a series of shielded
twisted-pair sections linking each station to its immediate
neighbors.
 Each station connects an output port on one section to an
input port on the next, creating a ring with one direction
traffic flow.
 A frame is passed to each station in sequence, where it is
examined, regenerated, and then sent on to the next station.
Note



If a node is disabled, it could stop the flow of traffic around
the entire network.
To solve the problem, each station is connected to an
automatic switch. The switch can bypass an inactive station.
Individual automatic switches are combined into a hub
called a (MAU) multi station access unit, which can
support up to 8 stations.
30
Figure 12-19
Token Ring Switch
31
Figure 12-20
MAU
32



A token ring network consists of a logical ring implemented in a physical
ring topology.
A token, which is a special frame, and data are transmitted in a point-topoint manner from one node (called a lobe) to the next.
The direction of circulation is fixed and either clockwise or
counterclockwise (but not both). E.g, on a counterclockwise rotating ring,
if lobe 3 has a “free” token & wants to send data to lobe 2, data frames
must circulate the ring in the order 3-4-5-1-2. On a clockwise rotating ring,
though, the transmission order is 3-2.
33
(a) A typical token ring network consists of lobes connected to a hub in
a physical star configuration
(b) Internally, lobes are actually interconnected via a logical ring
34
Example of a typical token ring network configuration.

Token ring hubs are called multistation access units (MAUs), and nodes are called
lobes.

Physically, Nodes are connected to an MAU in a star configuration. Within an MAU,
however, a logical ring topology exists.

Lobes are connected to the ring using an IBM Data Connector, which enables lobes to
be removed without disrupting the ring. MAUs also can be interconnected using special
“ring in/ring out” ports, which preserve the ring structure. Note the presence of relay
switches within each hub. Relay switches (also called bypass switches) are used to
maintain the integrity of the ring in the event of Node failure.
E.g: if Node 12 stops working or if there is a break in the cable connecting Node 12 to the
ring, the ring is broken. In such instances, the relay switch closes, thus preserving the
ring.
35
Format and contents of an IEEE 802.5 token frame.
36
Note that the S, AC, and E fields comprise the token frame
37

Incorporated within the backbone, a token ring switch acts as a
multiport source routing bridge and enables large networks to be
partitioned into smaller segments.
38
39
Fiber Distributed Data Interface
 FDDI Is a LAN protocol implemented by ANSI.
 It supports data rates of 100 Mbps
 When FDDI was designed, speeds of 100 Mbps required
fiber-optic cable.
 Today, however, this speed is available using copper cable.
The copper cable of FDDI is known as CDDI.
40
FDDI Rings
 Is implemented as a dual ring (primary & secondary)
 secondary is provided in case the primary fails.
 Whenever a problem occurs on the primary ring, the
secondary can be activated to complete the data circuits
and maintain service.
 Nodes connect to both rings using (MIC) media
Interface connector
41
FDDI Nodes
42
•FDDI’s counter-rotating ring architecture.
•The primary ring is active in normal operation; the
secondary ring provides redundancy.
•All devices on the ring are:
dual-attachment stations or dual-attachment hubs.
43
FDDI’s “self-healing” capability.
In the event of a fiber cut or an inoperative node, a FDDI
network automatically “heals” itself by wrapping the ring at
the point of failure. This is done by interconnecting the
primary and secondary rings into a single functional ring.
44
Access Method
1. Token passing, and is limited by time.
2. A station may send as many frames as it can
within its allotted access period, given that realtime data sent first.
45
FDDI defines 3 time registers to
control circulation of the token
Values are set when the ring is initialized.
1. Synchronous Allocation (SA):
This register indicates the length of time allowed each station for
sending synchronous data (real-Time). Which is different for each
station.
2. Target Token Rotation Time; (TTRT)
Indicates the average time required for a token to circulate around
the ring exactly once.
3. Absolute Maximum Time (AMT);
This register holds a value = 2*TTRT. Token may not take longer
than this time to make one rotation of the ring. If it does, the ring
must be initialized.
46
FDDI Frames
47
FDDI defines 3 types of nodes
1. Dual attachment station (DAS)


has two MIC (media Interface connector)
requires NIC with 2 inputs & 2 out puts
 Faults are bypassed by station’s making
connection from the primary ring to secondary to
switch signals from one input to another output
48
2.
Single attachment station (SAS)




SAS has 1 MIC (media Interface connector)
can connect only one ring.
Robustness is achieved by connecting SAS to (DACs),
rather than to the FDDI ring directly.
This configuration allows each workstation to operate
through a NIC with only one input and one output. The
connector (DAC) provides the connection to the dual ring.
49
Standards
 Different architectures could restrict the growth of
networking.
 The Institute of Electrical and Electronic Engineers (IEEE)
developed computer network architecture standards.
 The IEEE efforts were called Project 802.
 There are three dominant standards



Ethernet (802.3)
Token ring (802.5)
Wireless (802.11).
50
Ethernet
 Ethernet is the most popular LAN architecture today.
 It is easy to install and inexpensive.
 Data transmission is broken into packets.

10Base2




10BaseT




It runs 10 Mbps and uses CSMA/CD.
It is a star-wired bus topology.
It is still used today due to its reliability and ease of use
100BaseT




It is a bus topology that uses contention (CSMA/CD) and thin coax.
Segments are connected together through a repeater.
It can connect up to five 200-meter segments using four repeaters (5-4-3 rule)
It uses CMSA/CD as a star-wired bus.
There are three subcategories of 100Base-T networks with different cable requirements.
The most common is 100Base-TX (Category 5 or higher UTP).
Switched Ethernet

A switch knows which segment belongs to which device.
It uses a table stored in memory to send a packet
Gigabit Ethernet





1000Base-T is a star topology that uses Category 5 or higher cabling.
It increases speed by sending more bits and using 4 pairs of wires simultaneously.
10G Ethernet



It can be used in both LANs and WANs.
It requires fiber optic cable
Networks do not encounter collision
51
Token Ring




IEEE Project 802.5
16-Mbps speed
It uses twisted pair cable in a hybrid star ring topology
The packets go to a central hub called the Multistation
Access Unit (MAU).
52
WLAN
 Wireless local area network
 It is used when mobility is needed, but must remain
connected to the network.
 802.11b or Wi-Fi (11 Mbps)
 802.11a or Wi-Fi5 (108 Mbps)
 802.11g (54 Mbps)
53
WLAN
 Only wireless network interface cards and access points are needed.
 An access point acts as a link between wireless and wired networks.
 802.11b uses Carrier Sense Multiple Access with Collision Avoidance
(CSMA/CA).
54