03 The OSI Data Link Layer

Revision no.
: PPT/2K403/02
The OSI Data Link Layer

Revision no.: PPT/2K403/02
The OSI Data Link Layer
• Data Link layer protocol implementations do the following:
• Organize the Physical layer’s bits (1s and 0s) into logical
groups of information called frames. (Like a byte, a frame is a
contiguous series of bits grouped together as a unit of data.)
• Detect (and sometimes correct) errors.
• Control data flow.
• Identify computers on the network.
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
The OSI Data Link Layer (contd.)
The OSI Data Link Layer (contd.)
• The following networking connectivity devices are normally

associated with the OSI Data Link layer:
• Bridges
• Switches
• Network interface boards
OSI data link reference model:
Two sublayers
• Media access control. The media access control (MAC)

sublayer controls the way transmitters share a single
transmission channel.
• Logical link control. The logical link control (LLC) sublayer

establishes and maintains the link for transmitting data frames
from one device to the next.
Logical Topology Methods
Logical Topology Methods (contd.)
• The actual signal path is called a logical topology. Media

connection devices or media access schemes can be used to
pass a signal in a logical path that is different than the physical
media path.
• A good example of disparate physical and logical topologies is

a token ring network. Token ring LANs often use copper cable
arranged in a physical star topology with a hub at the center.
Logical Topology Versus Physical Topology
Logical Topology Versus Physical Topology (contd.)
• The hub does not repeat incoming signals to all other attached
devices. Instead, it sends each signal to the next device in a
predetermined logical ring or round-robin scheme. Therefore,
the physical topology is a star, while the logical topology is a
ring.
• Another good example is Ethernet, which can be wired as a

physical star but operate as a logical bus.
Media Access Methods
Media Access Methods (contd.)
• Logical topologies use specific rules that control when

network entities are allowed to transmit data signals. The
control process is called media access.
• Networks require some method of reducing collisions. The
following media access methods describe rules that govern
when network devices are allowed to transmit :
• Contention
• Token-passing
• Polling
Contention
Contention (contd.)
• Contention-based access methods allow network devices to

transmit data whenever they want, regardless of other devices
on the network.
• This scheme is simple and provides equal access rights to all

workstations.
Token-Passing Systems
Token-Passing Systems (contd.)
• In token-passing systems, a small frame (the token) is passed

from one device to another. A token is a special message that
temporarily gives media control to the device holding the
token. Passing the token around distributes access control
among the network’s devices.
Token-Passing Versus Contention
Token-Passing Versus Contention (contd.)
• Contention nor Token-passing provides consistently better

performance than the other. The actual performance benefits
depend on the implementation, the amount of traffic, and the
nature of the data being transmitted.
• When the load is heavy (traffic levels are high), token-passing

access methods provide higher throughput than contention.
However, contention can outperform token-passing in lightly
loaded conditions, because it requires little overhead in
managing media access.
Polling
Polling Systems
• Polling is an access method that designates one device (called
a controller, primary, or master) as a media access
administrator.
• This device queries each of the other devices (referred to as
secondaries) in a predetermined order to determine whether
they have information to transmit.
Polling Systems (contd.)
• To get data from a secondary, the primary addresses a request

for data to the secondary, and then receives the data the
secondary sends. The primary then polls another secondary
and receives the data that secondary sends, and so forth.
• The protocol limits how much data each secondary can

transmit after a poll.
• Polling systems are ideal for networking time-sensitive

devices, such as automation equipment.
Addressing Methods
Addressing Methods (contd.)
• Computer network entities need some way to distinguish

between different devices on the network. This is done
through Addressing.
• This is similar to the addressing system used by the postal

service. Postal carriers use building numbers in addition to
street, city, and country names to deliver mail. Often, zip codes
are also used to shortcutthe name portions of different
addresses.
Transmission Synchronization Methods
Transmission Synchronization Methods (contd.)
• The following transmission synchronization methods

determine how networks coordinate frame transmission:
• Asynchronous
• Synchronous
• Isochronous
Asynchronous Transmission
Asynchronous Transmission (contd.)
• Asynchronous transmission methods rely on the transmitting

and receiving devices to maintain their own internal clocks.
The two devices use similar timing, but they do not
synchronize their clocks.
• Asynchronous devices send each frame separately, using the

asynchronous bit synchronization method. Each frame, or bit
string, begins with a start bit and is transmitted with complete
disregard to the timing of other transmissions.
Synchronous Transmission
Synchronous Transmission (contd.)
• Synchronous transmission methods require that the

communicating devices take responsibility for providing a
transmission (or framing) clock. This can be accomplished by
one of the following methods:
• Synchronized transmissions resist timing errors much better

than asynchronous because both transmitter and receiver use
the same clock.
Synchronous Transmission (contd.)
• Synchronous transmission is also a more efficient method of
transmitting large blocks of data because it eliminates a lot of
the overhead associated with asynchronous transmission.
• Synchronous transmission can perform error detection. A
method known as Cyclic Redundancy Check(CRC) is often
used.
Connection Services Methods
LLC-level flow control
• Flow control rules regulate how much data can be transmitted

within a specified time. LLC-level flow control allows slow
devices to turn down faster ones.
• Flow control regulates both end devices (such as computers)

and intermediary devices (such as routers).
Error control
• For example, monetary transaction software used by

automated teller machines (ATMs) should use connection-
oriented services because the transaction must be completed
once and only once.
Window Flow-Control
Window Flow-Control (contd.)
• Window flow control establishes a buffer or window into which

a set number of data frames can be placed.
Mapping the 802.x Protocol Suite

to the OSI Reference Model
IEEE 802.1
• IEEE 802.1 defines Physical and Data Link Layer standards for
allowing one IEEE 802 LAN workstation to communicate with
another workstation on a different LAN or WAN.
IEEE 802.2
• Institute of Electrical and Electronic Engineers (IEEE).
• IEEE 802.2 defines the Logical Link Control (LLC) sublayer of
the IEEE 802.x series. It is commonly used with IEEE 802.3,
802.4, 802.5, and 802.6 implementations.
• IEEE 802.2 adds header fields to identify which upper layer
protocol is used in the frame and which Network layer
processes the frame’s source and destination use.
Mapping IEEE 802.3

Protocols to the OSI Model
IEEE 802.3 Physical Layer Standards
IEEE 802.3
• IEEE 802.3 specifies a variety of Physical layer options,
including different signaling modes (baseband and
broadband), media types, topologies, and data rates.
• The common element of each option is their use of the
CSMA/CD media access method.
IEEE 802.4
• IEEE 802.4 was created, primarily, to satisfy the LAN needs of
factory and industrial automation.
• This standard defines a physical bus topology, a token-
passing media access method, both baseband and broadband
media, and 75 ohm CATV-type cable or optical fiber.
IEEE 802.5
• IEEE 802.5 is based on IBM’s token ring specification.
• It uses a token-passing media access method and supports

data rates of 1, 4, or 16 Mbps.
IEEE 802.6
• IEEE 802.6 uses a data transfer technology called Distributed
Queue Dual Bus (DQDB) that allows both synchronous and
asynchronous traffic for supporting voice, video, and data
transmissions.
IEEE 802.7
• IEEE 802.7 defines a standard for the design, installation, and

test parameters of broadband communication.
IEEE 802.8
• IEEE 802.8 defines a Fiber Optic Technical Advisory Group to

work with the 802 working groups on fiber optic technology
and standards.
IEEE 802.9
• IEEE 802.9, Isochronous Ethernet (also known as IsoEnet),
focuses on the integration of voice transmissions with data
transmissions and supports both sporadic and patterned
traffic.
IEEE 802.10
• IEEE 802.10 defines a standard for services, protocols, data

formats, and interfaces that securely exchange data using
encryption mechanisms.
IEEE 802.11
• IEEE 802.11 defines standards for wireless LAN

implementations like spread-spectrum and infrared devices.
IEEE 802.12
• IEEE 802.12 defines a 100 Mbps physical star topology

contention-based network standard known as 100VG-AnyLAN.
• 802.12 network devices contend for media access by signaling

a hub. When multiple simultaneous transmission requests
occur, the hub grants transmission rights to the highest
priority traffic first.
• IEEE 802.12 supports both Ethernet and token ring frame

types.
Ethernet Network Topologies
Ethernet Network Topologies (contd.)
• Unlike IEEE 802.3, which specifies numerous options for
Physical layer specification, Ethernet’s traditional topology
was a linear bus, but is now most frequently implemented as a
physical star.
• All Ethernet implementations use the CSMA/CD contention-
based logical topology.
Thick Coaxial Ethernet Implementation
Thick Coaxial Ethernet Implementation (contd.)
• 10BASE5 uses thick coaxial cabling. This
• implementation is alsoreferred to as Thicknet or Thick

Ethernet.
• Network interfaces connect to the network using external

Ethernet transceivers and an Attachment Unit Interface (AUI)
or transceiver drop cables. Transceivers attach to the cable by
a “vampire” tap,
• A thick coaxial cable (RG-8) is used.
Thin Coaxial Ethernet Implementation
Thin Coaxial Ethernet Implementation (contd.)
• 10BASE2 uses thin coaxial cabling. If you use thin coaxial

cabling (RG-58 A/U or C/U)
• The board must be set to use the internal transceiver.
• Each end of the cable must be terminated with a 50 ohm

terminator.
• Maximum segment length is 185 meters (1000 feet).
• The maximum number of nodes per segment is 30 (including

repeaters).
Unshielded Twisted-Pair Ethernet Cabling
Unshielded Twisted-Pair Ethernet Cabling (contd.)
• 10BASE-T uses unshielded twisted-pair (UTP) cabling. If you
use twisted-pair Ethernet cabling as illustrated in the following
figure, the following rules apply :
• The maximum segment length is 100 meters (328 feet).
• The maximum number of workstations is 1,024.
• The maximum number of repeaters between communicating
stations is 4.
Hierarchical Star Topology
Hierarchical Star Topology (contd.)
• A UTP implementation uses a hierarchical star topology,
Devices typically connect to wall-plate jacks by UTP cable.
Permanent cable drops run from the wall-plate to a punchdown
block.
• Connections are made to a patch panel. Concentrators act as
signal combiners and repeaters. These concentrators can be
cascaded to support large numbers of connections.
Token Ring Network Cabling
Token Ring Network Cabling (contd.)
• IEEE 802.5 describes a token-passing ring network patterned
after rings researched and developed commercially by IBM.
IEEE 802.5 is compatible with IBM’s token ring network, and
differs in relatively minor ways.
• IEEE 802.5 specifies standards for the Physical layer and MAC
sublayer.
Comparing Fiber Distributed

Data Interface to the OSI Model
Fiber Distributed Data Interface to the OSI Model
• Like IEE 802.3 and 802.5, FDDI assumes its services will
primarily be used by IEEE 802.2 (LLC sublayer specification),
but does not assume other upper-layer protocols.
• FDDI was created to fill the need for a high-bandwidth, secure
local-to-medium-area network specification.
FDDI Counter-Rotating Rings
FDDI Counter-Rotating Rings (contd.)
• FDDI networks consist of two counter-rotating rings, Traffic
flows in one direction on one ring and in the opposite direction
on the other. One of these rings is the primary ring; the other
is the secondary.
• Under normal conditions, traffic flows only on the primary ring;
the secondary ring is used for out-of-band management.
FDDI Star Configuration
FDDI Star Configuration (contd.)
• One FDDI ring can include no more than 1,000 workstations,

and the total cable length must not exceed 200 kilometers
(km).
• Five hundred workstations and 100 km of cable are cited as

the limits for an FDDI ring. This is because if there is a primary
ring failure, the new ring visits each workstation twice.
• The FDDI cable specification requires multi-mode fiber optic

cables with a core diameter of 62.5 micrometers.
FDDI Token-Ring Passing Operation
FDDI Token-Ring Passing Operation (contd.)
• FDDI always releases a new token at the end of transmitted

frames. When the next workstation gets the token, it can add a
frame to the ring.
• There might, therefore, be multiple frames on the network at

once. This process contributes to FDDI’s high data rate
specifications.
• FDDI uses beaconing to isolate serious failures such as breaks

in the ring.
Mapping PPP and SLIP to the OSI Reference Model
SLIP & PPP
• Serial Line IP (SLIP) protocol and its successor, the Point-to-
Point Protocol (PPP), were designed to provide dial-up
(telephone) network connections to TCP/IP networks.
SLIP
• SLIP was the first of these protocols and was developed to
providedial-up IP connections.
• SLIP is easy to implement, but cannot perform many functions
such as the simultaneous transfer of multiple protocols.
• SLIP also is not a strict standard; two SLIP implementations
will often not work together.
Point-to-Point Protocol (PPP)
• Point-to-Point Protocol (PPP) was designed by the Internet

Engineering Task Force (IETF) to address SLIP shortcomings.
PPP can perform the following functions:
• Dynamic IP addressing
• Support for multiple protocols on the same link (by using a

field that identifies the packet’s upper-layer protocol)
• Password login
• Error control
Mapping X.25 to the OSI Reference Model
Mapping the X.25 Protocol
• X.25 is normally used to offer a dedicated continuous link
between two geographically separated offices.
• X.25 is a specification for physically attaching a computer to a
packet-switched network and transmitting packets.
• The functions it defines are assigned to the following three
levels (which can be mapped to the first three OSI layers) :
Mapping the X.25 Protocol (contd.)
• Level 1 includes Physical layer connectivity rules, which are
actually specified by other standards like X.21, X.21bis, V.32,
and so on.
• Level 2 specifies a connection-oriented data path, which is
defined by the Link Access Procedures-Balanced (LAPB)
protocol.This roughly corresponds to the LLC sublayer of the
OSI model.
Mapping the X.25 Protocol (contd.)
• Level 3 defines how packets are passed between data terminal

equipment (DTE), such as computers, and data circuit-
terminating equipment (DCE), such as the network interface.
Switching and routing algorithms are not specified by X.25, so
each vendor can implement its own strategy
Mapping Frame Relay to the OSI Reference Model
Mapping Frame Relay
• Frame relay is only concerned with supporting the OSI
Physical and Data Link layer functions, Frame relay is a
packet-oriented protocol designed for data traffic only. It can
also serve as a data access protocol in a circuit-switched
environment.
• Data Link Connection Identifiers (DLCIs) distinguish separate
virtual circuits across connections.
Mapping Frame Relay (contd.)
• Frame relay works with X.25 and Integrated Services Digital
Network (ISDN). It can also be implemented as a high-
performance, single-line interface for LAN bridges and routers.
• Frame relay can be implemented by 56K, T-1, or T-3 lines.
Mapping ISDN to the OSI Reference Model
Mapping ISDN
• Integrated Services Digital Network (ISDN) defines standards
for integrating analog and digital transmissions using digital
telecommunications networks.
• ISDN converts an existing analog telephone network to one
that can transmit both digital and analog data.
Mapping ISDN (contd.)
• The ISDN bit pipe can offer several standard rate multiplexed
channels with the following names and basic specifications:
• Channel A: 4 KHz analog channel
• Channel B: 64 Kbps digital channel
• Channel C: 8 or 16 Kbps digital channel (normally used for out-
of-band signaling)
Mapping ATM to the OSI Reference Model
Mapping ATM to the OSI Reference Model (contd.)
• Asynchronous Transfer Mode (ATM) is one of the network
standards for B-ISDN and cell relay.
• ATM is normally considered a LAN and a WAN protocol.
• ATM operates at the Network and Data Link layers on top of
various Physical layers such as SONET or FDDI. ATM supports
a wide variety of upper-layer protocols.
ATM
• ATM was designed to be as independent as possible from
specific physical device requirements so faster processing
speeds are possible. It can operate over T-1 and T-3 links
using single-mode or multi-mode fiber, STP, or UTP cabling.
• The prespecified bit rates are 155 Mbps and 622 Mbps.
However, speeds are expected to reach 10 Gbps.
Design & Published by:

CMS Institute, Design & Development Centre, CMS House, Plot No. 91, Street No.7,
MIDC, Marol, Andheri (E), Mumbai –400093, Tel: 91-22-28216511, 28329198
Email: courseware.inst@cmail.cms.co.in
www.cmsinstitute.co.in

03 The OSI Data Link Layer

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

03 The OSI Data Link Layer

Uploaded by

Copyright:

Available Formats

Revision no.

The OSI Data Link Layer

The OSI Data Link Layer

• Data Link layer protocol implementations do the following:

• Detect (and sometimes correct) errors.

• Control data flow.

• Identify computers on the network.

The OSI Data Link Layer (contd.)

The OSI Data Link Layer (contd.)

• The following networking connectivity devices are normally

• Network interface boards

OSI data link reference model:

• Media access control. The media access control (MAC)

• Logical link control. The logical link control (LLC) sublayer

Logical Topology Methods

Logical Topology Methods (contd.)

• The actual signal path is called a logical topology. Media

• A good example of disparate physical and logical topologies is

Logical Topology Versus Physical Topology

Logical Topology Versus Physical Topology (contd.)

• Another good example is Ethernet, which can be wired as a

Media Access Methods

Media Access Methods (contd.)

• Logical topologies use specific rules that control when

• Contention-based access methods allow network devices to

• This scheme is simple and provides equal access rights to all

Token-Passing Systems (contd.)

• In token-passing systems, a small frame (the token) is passed

Token-Passing Versus Contention

Token-Passing Versus Contention (contd.)

• Contention nor Token-passing provides consistently better

• When the load is heavy (traffic levels are high), token-passing

• Polling is an access method that designates one device (called

a controller, primary, or master) as a media access

• This device queries each of the other devices (referred to as

secondaries) in a predetermined order to determine whether

they have information to transmit.

Polling Systems (contd.)

• To get data from a secondary, the primary addresses a request

• The protocol limits how much data each secondary can

• Polling systems are ideal for networking time-sensitive

Addressing Methods (contd.)

• Computer network entities need some way to distinguish

• This is similar to the addressing system used by the postal

Transmission Synchronization Methods

Transmission Synchronization Methods (contd.)

• The following transmission synchronization methods

Asynchronous Transmission (contd.)

• Asynchronous transmission methods rely on the transmitting

• Asynchronous devices send each frame separately, using the

Synchronous Transmission (contd.)

• Synchronous transmission methods require that the

• Synchronized transmissions resist timing errors much better

Synchronous Transmission (contd.)

• Synchronous transmission is also a more efficient method of

transmitting large blocks of data because it eliminates a lot of

the overhead associated with asynchronous transmission.

• Synchronous transmission can perform error detection. A

method known as Cyclic Redundancy Check(CRC) is often

Connection Services Methods

LLC-level flow control

• Flow control rules regulate how much data can be transmitted

• Flow control regulates both end devices (such as computers)

• For example, monetary transaction software used by