Transcript Chapter 11: Approaches to Networking

Chapter 11:
Approaches to Networking
Business Data Communications, 4e
LANs, WANs, and MANs
Ownership
WANs can be either public or private
LANs are usually privately owned
Capacity
LANs are usually higher capacity, to carry greater internal
communications load
Coverage
LANs are typically limited to a single location
WANs interconnect locations
Comparison of
Networking Options
Types of WANs
Circuit-switched (today’s lecture)
Packet-switched (Thursday’s lecture)
Circuit-Switching
Definition:
Communication in which a
dedicated communications path is established
between two devices through one or more
intermediate switching nodes
Dominant in both voice and data
communications today
e.g. PSTN is a circuit-switched network
Relatively inefficient (100% dedication even
without 100% utilization)
Circuit-Switching Stages
Circuit establishment
Transfer of information
point-to-point from endpoints to node
internal switching/multiplexing among nodes
Circuit disconnect
Circuit Establishment
Station requests connection from node
Node determines best route, sends message to
next link
Each subsequent node continues the
establishment of a path
Once nodes have established connection, test
message is sent to determine if receiver is
ready/able to accept message
Information Transfer
Point-to-point transfer from source to node
Internal switching and multiplexed transfer
from node to node
Point-to-point transfer from node to receiver
Usually a full-duplex connection throughout
Circuit Disconnect
When transfer is complete, one station
initiates termination
Signals must be propagated to all nodes used
in transit in order to free up resources
Public Switched Telephone
Network (PSTN)
Subscribers
Local loop
Connects subscriber to
local telco exchange
Exchanges
Telco switching centers
Also known as end
office
>19,000 in US
Trunks
Connections between
exchanges
Carry multiple voice
circuits using FDM or
synchronous TDM
Managed by IXCs
(inter-exchange carriers)
Digital Circuit-Switching Node
Circuit Switching Node:
Digital Switch
Provides transparent signal path between any
pair of attached devices
Typically full-duplex
Circuit-Switching Node:
Network Interface
Provides hardware and functions to connect
digital devices to switch
Analog devices can be connected if interface
includes CODEC functions
Typically full-duplex
Circuit-Switching Node:
Control Unit
Establishes on-demand connections
Maintains connection while needed
Breaks down connection on completion
Blocking/Nonblocking Networks
Blocking: network is unable to connect two
stations because all possible paths are already
in use
Nonblocking: permits all possible connection
requests because any two stations can be
connected
Switching Techniques
Space-Division
Switching
Developed for analog
environment, but has been
carried over into digital
communication
Requires separate physical
paths for each signal
connection
Uses metallic or
semiconductor “gates”
Time-Division
Switching
Used in digital
transmission
Utilizes multiplexing to
place all signals onto a
common transmission
path
Bus must have higher
data rate than individual
I/O lines
Routing in Circuit-Switched
Networks
Requires balancing efficiency and resiliency
Traditional circuit-switched model is
hierarchical, sometimes supplemented with
peer-to-peer trunks
Newer circuit-switched networks are
dynamically routed: all nodes are peer-topeer, making routing more complex
Alternate Routing
Possible routes between two end offices are
predefined
Originating switch selects the best route for
each call
Routing paths can be fixed (1 route) or
dynamic (multiple routes, selected based on
current and historical traffic)
Control Signaling
Manage the establishment, maintenance, and
termination of signal paths
Includes signaling from subscriber to
network, and signals within network
In-channel signaling uses the same channel
for control signals and calls
Common-channel signaling uses independent
channels for controls (SS7)
1st generation: narrowband
ISDNISDN
Basic Rate Interface (BRI)
two 64Kbps bearer channels + 16Kbps data channel
(2B+D) = 144 Kbps
circuit-switched
2nd generation: broadband ISDN (B-ISDN)
Primary Rate Interface (PRI)
twenty-three 64Kbps bearer channels + 64 data channel
(23B+D) = 1.536 Mbps
packet-switched network
development effort led to ATM/cell relay
Past Criticism of ISDN
“Innovations Subscribers Don’t Need” , “It
Still Doesn’t Network” , “It Still Does
Nothing”
Why so much criticism?
overhyping of services before delivery
high price of equipment
delay in implementing infrastructure
incompatibility between providers' equipment.
ISDN Principles
Support of voice and nonvoice using limited set of
standard facilities
Support for switched and nonswitched applications
Reliance on 64kbps connections
Intelligence in the networks
Layered protocol architecture (can be mapped onto
OSI model)
Variety of configurations
ISDN User Interface
“Pipe” to user’s premises has fixed capacity
Standard physical interface can be used for
voice, data, etc
Use of the pipe can be a variable mix of voice
and data, up to the capacity
User can be charged based on use rather than
time
ISDN Network Architecture
Physical path from user to office
subscriber loop, aka local loop
full-duplex
primarily twisted pair, but fiber use growing
Central office connecting subscriber loops
B channels: 64kbps
D channels: 16 or 64kbps
H channels: 384, 1536, or 1920 kbps
Channel
Basic userISDN
channel B
(aka
“bearer channel”)
Can carry digital voice, data, or mixture
Mixed data must have same destination
Four kinds of connections possible
Circuit-switched
Packet-switched
Frame mode
Semipermanent
ISDN D Channel
Carries signaling information using commonchannel signaling
call management
billing data
Allows B channels to be used more
efficiently
Can be used for packet switching
ISDN H Channel
Only available over primary interface
High speed rates
Used in ATM
ISDN Basic Access
Basic Rate Interface (BRI)
Two full-duplex 64kbps B channels
One full-duplex 16kbps D channel
Framing, synchronization, and overhead
bring total data rate to 192kbps
Can be supported by existing twisted pair
local loops
2B+D most common, but 1B+D available
ISDN Primary Access
Primary Rate Interface (PRI)
Used when greater capacity required
No international agreement on rates
US, Canada, Japan: 1.544mbps (= to T1)
Europe: 2.048mbps
Typically 23 64kbps B + 1 64kbps D
Fractional use of nB+D possible
Can be used to support H channels
Packet-Switching
Includes
X.25, ISDN, ATM Networks
and frame-relay
technologies
Data is broken into packets, each of which can
be routed separately
Advantages: better line efficiency, signals can
always be routed, prioritization option
Disadvantages: transmission delay in nodes,
variable delays can cause jitter, extra overhead
for packet addresses
Packet-Switching Techniques
Datagram
each packet treated independently and referred to
as a datagram
packets may take different routes, arrive out of
sequence
Virtual Circuit
preplanned route established for all packets
similar to circuit switching, but the circuit is not
dedicated
Packet-Switched Routing
Adaptive routing changes based on network
conditions
Factors influencing routing are failure and
congestion
Nodes must exchange information on
network status
Tradeoff between quality and amount of
overhead
Packet-Switched Congestion
Control
When line utilization is >80%, queue length
grows too quickly
Congestion control limits queue length to
avoid througput problems
Status information exchanged among nodes
Control signals regulate data flow using
interface protocols (usually X.25)
X.25 Interface Standard
ITU-T standard for interface between host and
packet-switched network
Physical level handles physical connection between
host and link to the node
Technically X.21, but other standards can be substituted,
including RS-232
Link level provides for reliable data transfer
Uses LAPB, which is a subset of HDLC
Packet level provides virtual circuits between
Virtual-Circuit Service
External virtual circuit: logical connection
between two stations on the network
Internal virtual circuit: specific preplanned
route through the network
X.25 usually has a 1:1 relationship between
external and internal circuits
In some cases, X.25 can be implemented as a
packet-switched network
WANs for Voice
Requires very small and nonvariable delays
for natural conversation--difficult to provide
this with packet-switching
As a result, the preferred method for voice
transmission is circuit-switching
Most businesses use public telephone
networks, but a few organizations have
implemented private voice networks
WANs for Data
Public packet-switched networks (X.25)
Private packet-switched networks
Leased lines between sites (non-switched)
Public circuit-switched networks
Private circuit-switched networks
(interconnected digital PBXs)
ISDN (integrated X.25 and traditional circuitswitching)
WAN Considerations
Nature of traffic
stream generally works best with dedicated
circuits
bursty better suited to packet-switching
Strategic and growth control--limited with
public networks
Reliability--greater with packet-switching
Security--greater with private networks