Module Overview
33.1 WAN
Technologies Overview
33.1.1 WAN
technology
33.1.2 WAN
devices
33.1.3 WAN
Standards
33.1.4 WAN
encapsulation
33.1.5 Packet and
circuit switching
33.1.6 WAN link
options
33.2 WAN
Technologies
33.2.1 Analog
dialup
33.2.2 ISDN
33.2.3 Leased
line
33.2.4 X.25
33.2.5 Frame
Relay
33.2.6 ATM
33.2.7 DSL
33.2.8 Cable
modem
33.3 WAN
Design
33.3.1 WAN
communication
33.3.2 Steps in
WAN design
33.3.3 How to
identify and select networking capabilities
33.3.4
Three-layer design model
33.3.5 Other
layered design models
33.3.6 Other WAN
design considerations
Module: Summary
Overview
As the enterprise
grows beyond a single location, it is necessary to interconnect the LANs in the
various branches to form a wide-area network (WAN). This module examines some
of the options available for these interconnections, the hardware needed to
implement them, and the terminology used to discuss them.
There are many
options currently available today for implementing WAN solutions. They differ
in technology, speed, and cost. Familiarity with these technologies is an
important part of network design and evaluation.
If all data
traffic in an enterprise is within a single building, a LAN meets the needs of
the organization. Buildings can be interconnected with high-speed data links to
form a campus LAN if data must flow between buildings on a single campus.
However, a WAN is needed to carry data if it must be transferred between geographically
separate locations. Individual remote access to the LAN and connection of the
LAN to the Internet are separate study topics, and will not be considered here.
Most students
will not have the opportunity to design a new WAN, but many will be involved in
designing additions and upgrades to existing WANs, and will be able to apply
the techniques learned in this module.
Students
completing this module should be able to:
- Differentiate between a LAN and
WAN
- Identify the devices used in a
WAN
- List WAN standards
- Describe WAN encapsulation
- Classify the various WAN link
options
- Differentiate between
packet-switched and circuit-switched WAN technologies
- Compare and contrast current
WAN technologies
- Describe equipment involved in
the implementation of various WAN services
- Recommend a WAN service to an
organization based on its needs
- Describe DSL and cable modem
connectivity basics
- Describe a methodical procedure
for designing WANs
- Compare and contrast WAN
topologies
- Compare and contrast WAN design
models
- Recommend a WAN design to an
organization based on its needs
33.1 WAN Technologies Overview
33.1.1 WAN technology
A WAN is a data
communications network that operates beyond the geographic scope of a LAN. One
primary difference between a WAN and a LAN is that a company or organization
must subscribe to an outside WAN service provider in order to use WAN carrier
network services. A WAN uses data links provided by carrier services to access
the Internet and connect the locations of an organization to each other, to
locations of other organizations, to external services, and to remote users.
WANs generally carry a variety of traffic types, such as voice, data, and
video. Telephone and data services are the most commonly used WAN services.
Devices on the
subscriber premises are called customer premises equipment (CPE). The subscriber owns the CPE or leases the CPE
from the service provider. A copper or fiber cable connects the CPE to the
service provider’s nearest exchange or central office (CO). This cabling is
often called the local loop, or "last-mile". A dialed call is
connected locally to other local loops, or non-locally through a trunk to a
primary center. It then goes to a sectional center and on to a regional or
international carrier center as the call travels to its destination.
In order for the
local loop to carry data, a device such as a modem is needed to prepare the
data for transmission. Devices that put data on the local loop are called data
circuit-terminating equipment, or data communications equipment (DCE). The
customer devices that pass the data to the DCE are called data terminal
equipment (DTE). The DCE primarily
provides an interface for the DTE into the communication link on the WAN cloud.
The DTE/DCE interface uses various physical layer protocols, such as High-Speed
Serial Interface (HSSI) and V.35. These protocols establish the codes and
electrical parameters the devices use to communicate with each other.
WAN links are provided at
various speeds measured in bits per second (bps), kilobits per second (kbps or
1000 bps), megabits per second (Mbps or 1000 kbps) or gigabits per second (Gbps
or 1000 Mbps). The bps values are generally full duplex. This means that an E1
line can carry 2 Mbps, or a T1 can carry 1.5 Mbps, in each direction
simultaneously.
33.1
WAN Technologies Overview
33.1.2 WAN devices
WANs are groups
of LANs connected together with communications links from a service provider.
Because the communications links cannot plug directly into the LAN, it is
necessary to identify the various pieces of interfacing equipment.
LAN-based
computers with data to transmit send data to a router that contains both LAN
and WAN interfaces. The router will use
the Layer 3 address information to deliver the data on the appropriate WAN
interface. Routers are active and intelligent network devices and therefore can
participate in network management. Routers manage networks by providing dynamic
control over resources and supporting the tasks and goals for networks. Some of
these goals are connectivity, reliable performance, management control, and
flexibility.
The
communications link needs signals in an appropriate format. For digital lines,
a channel service unit (CSU) and a data service unit (DSU) are required. The
two are often combined into a single piece of equipment, called the CSU/DSU.
The CSU/DSU may also be built into the interface card in the router.
A modem is needed
if the local loop is analog rather than digital. Modems transmit data over voice-grade
telephone lines by modulating and demodulating the signal. The digital signals
are superimposed on an analog voice signal that is modulated for transmission.
The modulated signal can be heard as a series of whistles by turning on the
internal modem speaker. At the receiving end the analog signals are returned to
their digital form, or demodulated.
When ISDN is used
as the communications link, all equipment attached to the ISDN bus must be
ISDN-compatible. Compatibility is generally built into the computer interface
for direct dial connections, or the router interface for LAN to WAN
connections. Older equipment without an ISDN interface requires an ISDN
terminal adapter (TA) for ISDN compatibility.
Communication
servers concentrate dial-in user communication and remote access to a LAN. They
may have a mixture of analog and digital (ISDN) interfaces and support hundreds
of simultaneous users.
33.1 WAN Technologies Overview
33.1.3 WAN Standards
WANs use the OSI
reference model, but focus mainly on Layer 1 and Layer 2. WAN standards
typically describe both physical layer delivery methods and data link layer
requirements, including physical addressing, flow control, and encapsulation.
WAN standards are defined and managed by a number of recognized authorities.
The physical
layer protocols describe how to provide electrical, mechanical, operational,
and functional connections to the services provided by a communications service
provider. Some of the common physical layer standards are listed in Figure ,
and their connectors illustrated in Figure .
The data link
layer protocols define how data is encapsulated for transmission to remote
sites, and the mechanisms for transferring the resulting frames. A variety of
different technologies are used, such as ISDN, Frame Relay or Asynchronous
Transfer Mode (ATM). These protocols use the same basic framing mechanism,
high-level data link control (HDLC), an ISO standard, or one of its sub-sets or
variants.
33.1 WAN Technologies Overview
33.1.4
WAN encapsulation
Data from the
network layer is passed to the data link layer for delivery on a physical link,
which is normally point-to-point on a WAN connection. The data link layer
builds a frame around the network layer data so the necessary checks and
controls can be applied. Each WAN connection type uses a Layer 2 protocol to
encapsulate traffic while it is crossing the WAN link. To ensure that the
correct encapsulation protocol is used, the Layer 2 encapsulation type used for
each router serial interface must be configured. The choice of encapsulation
protocols depends on the WAN technology and the equipment. Most framing is
based on the HDLC standard.
HDLC framing
gives reliable delivery of data over unreliable lines and includes signal
mechanisms for flow and error control.
The frame always starts and ends with an 8-bit flag field, the bit
pattern is 01111110. Because there is a likelihood that this pattern will occur
in the actual data, the sending HDLC system always inserts a 0 bit after every
five 1s in the data field, so in practice the flag sequence can only occur at
the frame ends. The receiving system strips out the inserted bits. When frames
are transmitted consecutively the end flag of the first frame is used as the start
flag of the next frame.
The address field
is not needed for WAN links, which are almost always point-to-point. The
address field is still present and may be one or two bytes long. The control
field indicates the frame type, which may be information, supervisory, or
unnumbered:
- Unnumbered frames carry line
setup messages.
- Information frames carry
network layer data.
- Supervisory frames control the
flow of information frames and request data retransmission in the event of
an error.
The control field is normally
one byte, but will be two bytes for extended sliding windows systems. Together
the address and control fields are called the frame header. The encapsulated
data follows the control field. Then a frame check sequence (FCS) uses the
cyclic redundancy check (CRC) mechanism to establish a two or four byte field.
Several data link
protocols are used, including sub-sets and proprietary versions of HDLC. Both PPP and the Cisco version of HDLC have
an extra field in the header to identify the network layer protocol of the
encapsulated data.
33.1
WAN Technologies Overview
33.1.5 Packet and circuit switching
Packet-switched
networks were developed to overcome the expense of public circuit-switched
networks and to provide a more cost-effective WAN technology.
When a subscriber
makes a telephone call, the dialed number is used to set switches in the
exchanges along the route of the call so that there is a continuous circuit
from the originating caller to that of the called party. Because of the
switching operation used to establish the circuit, the telephone system is
called a circuit-switched network. If the telephones are replaced with modems,
then the switched circuit is able to carry computer data.
The internal path
taken by the circuit between exchanges is shared by a number of conversations.
Time division multiplexing (TDM) is used to give each conversation a share of
the connection in turn. TDM assures that a fixed capacity connection is made
available to the subscriber.
If the circuit
carries computer data, the usage of this fixed capacity may not be efficient.
For example, if the circuit is used to access the Internet, there will be a
burst of activity on the circuit while a web page is transferred. This could be
followed by no activity while the user reads the page and then another burst of
activity while the next page is transferred. This variation in usage between
none and maximum is typical of computer network traffic. Because the subscriber
has sole use of the fixed capacity allocation, switched circuits are generally
an expensive way of moving data.
An alternative is
to allocate the capacity to the traffic only when it is needed, and share the
available capacity between many users. With a circuit-switched connection, the
data bits put on the circuit are automatically delivered to the far end because
the circuit is already established. If the circuit is to be shared, there must
be some mechanism to label the bits so that the system knows where to deliver
them. It is difficult to label individual bits, therefore they are gathered
into groups called cells, frames, or packets. The packet passes from exchange
to exchange for delivery through the provider network. Networks that implement
this system are called packet-switched networks.
The links that
connect the switches in the provider network belong to an individual subscriber
during data transfer, therefore many subscribers can share the link. Costs can
be significantly lower than a dedicated circuit-switched connection. Data on
packet-switched networks are subject to unpredictable delays when individual
packets wait for other subscriber packets to be transmitted by a switch.
The switches in a
packet-switched network determine, from addressing information in each packet,
which link the packet must be sent on next. There are two approaches to this
link determination, connectionless or connection-oriented. Connectionless
systems, such as the Internet, carry full addressing information in each
packet. Each switch must evaluate the address to determine where to send the
packet. Connection-oriented systems predetermine the route for a packet, and
each packet need only carry an identifier. In the case of Frame Relay, these
are called Data Link Control Identifiers (DLCI). The switch determines the
onward route by looking up the identifier in tables held in memory. The set of
entries in the tables identifies a particular route or circuit through the
system. If this circuit is only physically in existence while a packet is
traveling through it, it is called a Virtual Circuit (VC).
The table entries
that constitute a VC can be established by sending a connection request through
the network. In this case the resulting circuit is called a Switched Virtual
Circuit (SVC). Data that is to travel on SVCs must wait until the table entries
have been set up. Once established, the SVC may be in operation for hours, days
or weeks. Where a circuit is required to be always available, a Permanent
Virtual Circuit (PVC) will be established. Table entries are loaded by the
switches at boot time so the PVC is always available.
33.1
WAN Technologies Overview
33.1.6 WAN link options
Figure provides an overview of WAN link options.
Circuit switching
establishes a dedicated physical connection for voice or data between a sender
and receiver. Before communication can start, it is necessary to establish the
connection by setting the switches. This is done by the telephone system, using
the dialed number. ISDN is used on digital lines as well as on voice-grade
lines.
To avoid the
delays associated with setting up a connection, telephone service providers
also offer permanent circuits. These dedicated or leased lines offer higher
bandwidth than is available with a switched circuit. Examples of circuit-switched
connections include:
- Plain Old Telephone System
(POTS)
- ISDN Basic Rate Interface (BRI)
- ISDN Primary Rate Interface
(PRI)
Many WAN users do
not make efficient use of the fixed bandwidth that is available with dedicated,
switched, or permanent circuits, because the data flow fluctuates.
Communications providers have data networks available to more appropriately
service these users. In these networks, the data is transmitted in labeled
cells, frames, or packets through a packet-switched network. Because the
internal links between the switches are shared between many users, the costs of
packet switching are lower than those of circuit switching. Delays (latency)
and variability of delay (jitter) are greater in packet-switched than in
circuit-switched networks. This is because the links are shared and packets
must be entirely received at one switch before moving to the next. Despite the
latency and jitter inherent in shared networks, modern technology allows
satisfactory transport of voice and even video communications on these
networks.
Packet-switched
networks may establish routes through the switches for particular end-to-end
connections. Routes established when the switches are started are PVCs. Routes
established on demand are SVCs. If the routing is not pre-established and is
worked out by each switch for each packet, the network is called
connectionless.
To connect to a
packet-switched network, a subscriber needs a local loop to the nearest
location where the provider makes the service available. This is called the
point-of-presence (POP) of the service. Normally this will be a dedicated
leased line. This line will be much shorter than a leased line directly
connected to the subscriber locations, and often carries several VCs. Since it is likely that not all the VCs will
require maximum demand simultaneously, the capacity of the leased line can be
smaller than the sum of the individual VCs. Examples of packet or cell switched
connections include:
- Frame Relay
- X.25
- ATM
33.2 WAN Technologies
33.2.1 Analog dialup
When
intermittent, low-volume data transfers are needed, modems and analog dialed
telephone lines provide low capacity and dedicated switched connections.
Traditional
telephony uses a copper cable, called the local loop, to connect the telephone
handset in the subscriber premises to the public switched telephone network
(PSTN). The signal on the local loop during a call is a continuously varying
electronic signal that is a translation of the subscriber voice.
The local loop is
not suitable for direct transport of binary computer data, but a modem can send
computer data through the voice telephone network. The modem modulates the
binary data into an analog signal at the source and demodulates the analog
signal at the destination to binary data.
The physical
characteristics of the local loop and its connection to the PSTN limit the rate
of the signal. The upper limit is around 33 kbps. The rate can be increased to
around 56 kbps if the signal is coming directly through a digital connection.
For small
businesses, this can be adequate for the exchange of sales figures, prices,
routine reports, and email. Using automatic dialup at night or on weekends for
large file transfers and data backup can take advantage of lower off-peak
tariffs (line charges). Tariffs are based on the distance between the
endpoints, time of day, and the duration of the call.
The advantages of
modem and analog lines are simplicity, availability, and low implementation
cost. The disadvantages are the low data rates and a relatively long connection
time. The dedicated circuit provided by dialup will have little delay or jitter
for point-to-point traffic, but voice or video traffic will not operate
adequately at relatively low bit rates.
33.2 WAN Technologies
33.2.2 ISDN
The internal
connections, or trunks, of the PSTN have changed from carrying analog
frequency-division multiplexed signals, to time-division multiplexed (TDM)
digital signals. An obvious next step is to enable the local loop to carry
digital signals that result in higher capacity switched connections.
Integrated
Services Digital Network (ISDN) turns the local loop into a TDM digital
connection. The connection uses 64 kbps bearer channels (B) for carrying voice
or data and a signaling, delta channel (D) for call set-up and other purposes.
Basic Rate
Interface (BRI) ISDN is intended for the home and small enterprise and provides
two 64 kbps B channels and a 16 kbps D channel. For larger installations,
Primary Rate Interface (PRI) ISDN is available. PRI delivers twenty-three 64
kbps B channels and one 64 kbps D channel in North America, for a total bit
rate of up to 1.544 Mbps. This includes some additional overhead for
synchronization. In Europe, Australia, and other parts of the world, ISDN PRI
provides thirty B channels and one D channel for a total bit rate of up to
2.048 Mbps, including synchronization overhead.
In North America PRI corresponds to a T1 connection. The rate of
international PRI corresponds to an E1 connection.
The BRI D channel
is underutilized, as it has only two B channels to control. Some providers
allow the D channel to carry data at low bit rates such as X.25 connections at
9.6 kbps.
For small WANs,
the BRI ISDN can provide an ideal connection mechanism. BRI has a call setup
time that is less than a second, and its 64 kbps B channel provide greater
capacity than an analog modem link. If
greater capacity is required, a second B channel can be activated to provide a total
of 128 kbps. Although inadequate for video, this would permit several
simultaneous voice conversations in addition to data traffic.
Another common
application of ISDN is to provide additional capacity as needed on a leased
line connection. The leased line is sized to carry average traffic loads while
ISDN is added during peak demand periods. ISDN is also used as a backup in the
case of a failure of the leased line. ISDN tariffs are based on a per-B channel
basis and are similar to those of analog voice connections.
With PRI ISDN,
multiple B channels can be connected between two end points. This allows for
video conferencing and high bandwidth data connections with no latency or
jitter. Multiple connections can become very expensive over long distances.
33.2
WAN Technologies
33.2.3 Leased line
When permanent
dedicated connections are required, leased lines are used with capacities
ranging up to 2.5 Gbps.
A point-to-point
link provides a pre-established WAN communications path from the customer
premises through the provider network to a remote destination. Point-to-point
lines are usually leased from a carrier and are called leased lines. Leased
lines are available in different capacities.
These dedicated circuits are generally priced based on bandwidth
required and distance between the two connected points. Point-to-point links
are generally more expensive than shared services such as Frame Relay. The cost
of leased-line solutions can become significant when they are used to connect
many sites. There are times when cost of the leased line is outweighed by the
benefits. The dedicated capacity gives no latency or jitter between the
endpoints. Constant availability is essential for some applications such as
electronic commerce.
A router serial
port is required for each leased-line connection. A CSU/DSU and the actual
circuit from the service provider are also required.
Leased lines are
used extensively for building WANs and give permanent dedicated capacity. They have been the traditional connection of
choice but have a number of disadvantages. WAN traffic is often variable and
leased lines have a fixed capacity. This results in the bandwidth of the line
seldom being exactly what is needed. In addition, each end point would need an
interface on the router which would increase equipment costs. Any changes to
the leased line generally require a site visit by the carrier to change
capacity.
Leased lines
provide direct point-to-point connections between enterprise LANs and connect
individual branches to a packet-switched network. Several connections can be
multiplexed over a leased line, resulting in shorter links and fewer required
interfaces.
33.2 WAN Technologies
33.2.4 X.25
In response to
the expense of leased lines, telecommunications providers introduced packet-switched
networks using shared lines to reduce costs. The first of these packet-switched
networks was standardized as the X.25 group of protocols. X.25 provides a low
bit rate shared variable capacity that may be either switched or permanent.
X.25 is a network-layer
protocol and subscribers are provided with a network address. Virtual circuits
can be established through the network with call request packets to the target
address. The resulting SVC is identified by a channel number. Data packets
labeled with the channel number are delivered to the corresponding address.
Multiple channels can be active on a single connection.
Subscribers
connect to the X.25 network with either leased lines or dialup connections.
X.25 networks can also have pre-established channels between subscribers that
provide a PVC.
X.25 can be very
cost effective because tariffs are based on the amount of data delivered rather
than connection time or distance. Data can be delivered at any rate up to the
connection capacity. This provides some flexibility. X.25 networks are usually
low capacity, with a maximum of 48 kbps. In addition, the data packets are
subject to the delays typical of shared networks.
X.25 technology
is no longer widely available as a WAN technology in the US. Frame Relay has
replaced X.25 at many service provider locations.
Typical X.25
applications are point-of-sale card readers. These readers use X.25 in dialup
mode to validate transactions on a central computer. Some enterprises also use
X.25 based value-added networks (VAN) to transfer Electronic Data Interchange
(EDI) invoices, bills of lading, and other commercial documents. For these
applications, the low bandwidth and high latency are not a concern, because the
low cost makes the use of X.25 affordable.
33.2 WAN Technologies
33.2.5 Frame Relay
With increasing
demand for higher bandwidth and lower latency packet switching, communications
providers introduced Frame Relay. Although the network layout appears similar
to that for X.25, available data rates are commonly up to 4 Mbps, with some
providers offering even higher rates.
Frame Relay
differs from X.25 in several aspects. Most importantly, it is a much simpler
protocol that works at the data link layer rather than the network layer.
Frame Relay implements
no error or flow control. The simplified handling of frames leads to reduced
latency, and measures taken to avoid frame build-up at intermediate switches
help reduce jitter.
Most Frame Relay
connections are PVCs rather than SVCs. The connection to the network edge is
often a leased line but dialup connections are available from some providers
using ISDN lines. The ISDN D channel is used to set up an SVC on one or more B
channels. Frame Relay tariffs are based on the capacity of the connecting port
at the network edge. Additional factors are the agreed capacity and committed
information rate (CIR) of the various PVCs through the port.
Frame Relay
provides permanent shared medium bandwidth connectivity that carries both voice
and data traffic. Frame Relay is ideal for connecting enterprise LANs. The
router on the LAN needs only a single interface, even when multiple VCs are
used. The short-leased line to the Frame Relay network edge allows
cost-effective connections between widely scattered LANs.
33.2
WAN Technologies
33.2.6 ATM
Communications
providers saw a need for a permanent shared network technology that offered
very low latency and jitter at much higher bandwidths. Their solution was
Asynchronous Transfer Mode (ATM). ATM has data rates beyond 155 Mbps. As with
the other shared technologies, such as X.25 and Frame Relay, diagrams for ATM
WANs look the same.
ATM is a
technology that is capable of transferring voice, video, and data through
private and public networks. It is built on a cell-based architecture rather
than on a frame-based architecture. ATM cells are always a fixed length of 53
bytes. The 53 byte ATM cell contains a 5 byte ATM header followed by 48 bytes
of ATM payload. Small, fixed-length cells are well suited for carrying voice
and video traffic because this traffic is intolerant of delay. Video and voice
traffic do not have to wait for a larger data packet to be transmitted.
The 53 byte ATM
cell is less efficient than the bigger frames and packets of Frame Relay and
X.25. Furthermore, the ATM cell has at least 5 bytes of overhead for each
48-byte payload. When the cell is carrying segmented network layer packets, the
overhead will be higher because the ATM switch must be able to reassemble the
packets at the destination. A typical ATM line needs almost 20% greater
bandwidth than Frame Relay to carry the same volume of network layer data.
ATM offers both
PVCs and SVCs, although PVCs are more common with WANs.
As with other
shared technologies, ATM allows multiple virtual circuits on a single leased
line connection to the network edge.
33.2
WAN Technologies
33.2.7 DSL
Digital
Subscriber Line (DSL) technology is a broadband technology that uses existing
twisted-pair telephone lines to transport high-bandwidth data to service
subscribers. DSL service is considered broadband, as opposed to the baseband
service for typical LANs. Broadband refers to a technique which uses multiple
frequencies within the same physical medium to transmit data. The term xDSL covers
a number of similar yet competing forms of DSL technologies:
- Asymmetric DSL (ADSL)
- Symmetric DSL (SDSL)
- High Bit Rate DSL (HDSL)
- ISDN (like) DSL (IDSL)
- Consumer DSL (CDSL), also
called DSL-lite or G.lite
DSL technology
allows the service provider to offer high-speed network services to customers,
utilizing installed local loop copper lines. DSL technology allows the local
loop line to be used for normal telephone voice connection and an always-on
connection for instant network connectivity. Multiple DSL subscriber lines are
multiplexed into a single, high capacity link by the use of a DSL Access
Multiplexer (DSLAM) at the provider location. DSLAMs incorporate TDM technology
to aggregate many subscriber lines into a less cumbersome single medium, generally
a T3/DS3 connection. Current DSL technologies are using sophisticated coding
and modulation techniques to achieve data rates up to 8.192 Mbps.
The voice channel
of a standard consumer telephone covers the frequency range of 330 Hz to 3.3
KHz. A frequency range, or window, of 4 KHz is regarded as the requirements for
any voice transmission on the local loop. DSL technologies place upload
(upstream) and download (downstream) data transmissions at frequencies above
this 4 KHz window. This technique is what allows both voice and data
transmissions to occur simultaneously on a DSL service.
The two basic
types of DSL technologies are asymmetric (ADSL) and symmetric (SDSL). All forms
of DSL service are categorized as ADSL or SDSL and there are several varieties
of each type. Asymmetric service provides higher download or downstream
bandwidth to the user than upload bandwidth. Symmetric service provides the
same capacity in both directions.
Not all DSL
technologies allow the use of a telephone. SDSL is called dry copper because it
does not have a ring tone and does not offer telephone service on the same
line. Therefore a separate line is required for the SDSL service.
The different
varieties of DSL provide different bandwidths, with capabilities exceeding
those of a T1 or E1 leased line. The transfer rates are dependent on the actual
length of the local loop and the type and condition of its cabling. For
satisfactory service, the loop must be less than 5.5 kilometers (3.5 miles).
DSL availability is far from universal, and there are a wide variety of types,
standards, and emerging standards. It is not a popular choice for enterprise
computer departments to support home workers. Generally, a subscriber cannot
choose to connect to the enterprise network directly, but must first connect to
an Internet service provider (ISP). From here, an IP connection is made through
the Internet to the enterprise. Thus, security risks are incurred. To address
security concerns, DSL services provide capabilities for using Virtual Private
Network (VPN) connections to a VPN server, which is typically located at the
corporate site.
33.2
WAN Technologies
33.2.8 Cable modem
Coaxial cable is
widely used in urban areas to distribute television signals. Network access is available from some cable
television networks. This allows for greater bandwidth than the conventional
telephone local loop.
Enhanced cable
modems enable two-way, high-speed data transmissions using the same coaxial
lines that transmit cable television. Some cable service providers are
promising data speeds up to 6.5 times that of T1 leased lines. This speed makes
cable an attractive medium for transferring large amounts of digital
information quickly, including video clips, audio files, and large amounts of
data. Information that would take two minutes to download using ISDN BRI can be
downloaded in two seconds through a cable modem connection.
Cable modems
provide an always-on connection and a simple installation. An always-on cable
connection means that connected computers are vulnerable to a security breach
at all times and need to be suitably secured with firewalls. To address
security concerns, cable modem services provide capabilities for using Virtual
Private Network (VPN) connections to a VPN server, which is typically located
at the corporate site.
A cable modem is
capable of delivering up to 30 to 40 Mbps of data on one 6 MHz cable channel.
This is almost 500 times faster than a 56 Kbps modem.
With a cable
modem, a subscriber can continue to receive cable television service while
simultaneously receiving data to a personal computer. This is accomplished with
the help of a simple one-to-two splitter.
Cable modem
subscribers must use the ISP associated with the service provider. All the
local subscribers share the same cable bandwidth. As more users join the
service, available bandwidth may be below the expected rate. -
33.3 WAN Design
33.3.1 WAN communication
WANS are
considered to be a set of data links connecting routers on LANs. User end
stations and servers on LANs exchange data. Routers pass data between networks
across the data links.
Because of cost
and legal reasons, a communications provider or a common carrier normally owns
the data links that make up a WAN. The links are made available to subscribers
for a fee and are used to interconnect LANs or connect to remote networks. WAN
data transfer speed (bandwidth) is considerably slower than the 100 Mbps that
is common on a LAN. The charges for link provision are the major cost element
of a WAN and the design must aim to provide maximum bandwidth at acceptable
cost. With user pressure to provide more service access at higher speeds and
management pressure to contain cost, determining the optimal WAN configuration
is not an easy task.
WANs carry a
variety of traffic types such as data, voice, and video. The design selected
must provide adequate capacity and transit times to meet the requirements of
the enterprise. Among other specifications, the design must consider the
topology of the connections between the various sites, the nature of those
connections, and bandwidth capacity.
Older WANs often
consisted of data links directly connecting remote mainframe computers. Today’s WANs, though, connect geographically
separated LANs. End-user stations,
servers, and routers communicate across LANs, and the WAN data links terminate
at local routers. By exchanging Layer 3 address information about directly
connected LANs, routers determine the most appropriate path through the network
for the required data streams. Routers can also provide quality of service
(QoS) management, which allots priorities to the different traffic streams.
Because the WAN
is merely a set of interconnections between LAN based routers, there are no
services on the WAN. WAN technologies function at the lower three layers of the
OSI reference model. Routers determine
the destination of the data from the network layer headers and transfer the
packets to the appropriate data link connection for delivery on the physical
connection.
33.3
WAN Design
33.3.2 Steps in WAN design
Designing a WAN
can be a challenging task, but approaching the design in a systematic manner
can lead to superior performance at a reduced cost. Many WANs have evolved over
time, therefore many of the guidelines discussed here may not have been
considered. Every time a modification to an existing WAN is considered, the
steps in this module should be followed. WAN modifications may arise from
changes such as an expansion in the enterprise the WAN serves, or accommodation
of new work practices and business methods.
Enterprises
install WAN connectivity because there is a need to move data in a timely
manner between external branches. The WAN is there to support the enterprise
requirements. Meeting these requirements incurs costs, such as equipment
provisioning and management of the data links.
In designing the
WAN, it is necessary to know what data traffic must be carried, its origin, and
its destination. WANs carry a variety of traffic types with varying
requirements for bandwidth, latency, and jitter.
For each pair of
end points and for each traffic type, information is needed on the various
traffic characteristics. Determining
this may involve extensive studies of and consultation with the network users.
The design often involves upgrading, extending, or modifying an existing WAN.
Much of the data needed can come from existing network management statistics.
Knowing the
various end points allows the selection of a topology or layout for the WAN.
The topology will be influenced by geographic considerations but also by
requirements such as availability. A high requirement for availability will
require extra links that provide alternative data paths for redundancy and load
balancing.
With the end
points and the links chosen, the necessary bandwidth can be estimated. Traffic
on the links may have varying requirements for latency and jitter. With the
bandwidth availability already determined, suitable link technologies must be selected.
Finally,
installation and operational costs for the WAN can be determined and compared
with the business need driving the WAN provision.
In practice,
following the steps shown in Figure is
seldom a linear process. Several modifications may be necessary before a design
is finalized. Continued monitoring and re-evaluation are also required after
installation of the WAN to maintain optimal performance.
33.3
WAN Design
33.3.3
How to identify and select networking capabilities
Designing a WAN
essentially consists of the following:
- Selecting an interconnection
pattern or layout for the links between the various locations
- Selecting the technologies for
those links to meet the enterprise requirements at an acceptable cost
Many WANs use a star topology.
As the enterprise grows and new branches are added, the branches are connected
back to the head office, producing a traditional star topology. Star end-points are sometimes
cross-connected, creating a mesh or partial mesh topology. This provides for many possible combinations
for interconnections. When designing, re-evaluating, or modifying a WAN, a
topology that meets the design requirements must be selected.
In selecting a
layout, there are several factors to consider. More links will increase the
cost of the network services, and having multiple paths between destinations
increases reliability. Adding more network devices to the data path will
increase latency and decrease reliability. Generally, each packet must be
completely received at one node before it can be passed to the next. A range of
dedicated technologies with different features is available for the data links.
Technologies that
require the establishment of a connection before data can be transmitted, such
as basic telephone, ISDN, or X.25, are not suitable for WANs that require rapid
response time or low latency. Once established, ISDN and other dialup services
are low latency, low jitter circuits. ISDN is often the application of choice
for connecting a small office or home office (SOHO) network to the enterprise
network, providing reliable connectivity and adaptable bandwidth. Unlike cable
and DSL, ISDN is an option wherever modern telephone service is available. ISDN
is also useful as a backup link for primary connections and for providing
bandwidth-on-demand connections in parallel with a primary connection. A
feature of these technologies is that the enterprise is only charged a fee when
the circuit is in use.
The different
parts of the enterprise may be directly connected with leased lines, or they
may be connected with an access link to the nearest point-of-presence (POP) of
a shared network. X.25, Frame Relay, and ATM are examples of shared networks.
Leased lines will generally be much longer and therefore more expensive than
access links, but are available at virtually any bandwidth. They provide very
low latency and jitter.
ATM, Frame Relay,
and X.25 networks carry traffic from several customers over the same internal
links. The enterprise has no control over the number of links or hops that data
must traverse in the shared network. It cannot control the time data must wait
at each node before moving to the next link. This uncertainty in latency and
jitter makes these technologies unsuitable for some types of network traffic.
However, the disadvantages of a shared network may often be outweighed by the
reduced cost. Because several customers are sharing the link, the cost to each
will generally be less than the cost of a direct link of the same capacity.
Although ATM is a
shared network, it has been designed to produce minimal latency and jitter
through the use of high-speed internal links sending easily manageable units of
data, called cells. ATM cells have a fixed length of 53 bytes, 48 for data and
5 for the header. ATM is widely used for carrying delay-sensitive traffic.
Frame Relay may also be used for delay-sensitive traffic, often using QoS
mechanisms to give priority to the more sensitive data.
A typical WAN
uses a combination of technologies that are usually chosen based on traffic
type and volume. ISDN, DSL, Frame Relay,
or leased lines are used to connect individual branches into an area. Frame
Relay, ATM, or leased lines are used to connect external areas back to the
backbone. ATM or leased lines form the WAN backbone.
33.3
WAN Design
33.3.4 Three-layer design model
A systematic
approach is needed when many locations must be joined. A hierarchical solution
with three layers offers many advantages.
Imagine an
enterprise that is operational in every country of the European Union and has a
branch in every town with a population over 10,000. Each branch has a LAN, and
it has been decided to interconnect the branches. A mesh network is clearly not
feasible because nearly 500,000 links would be needed for the 900 centers. A
simple star will be very difficult to implement because it needs a router with
900 interfaces at the hub or a single interface that carries 900 virtual
circuits to a packet-switched network.
Instead, consider
a hierarchical design model. A group of LANs in an area are interconnected,
several areas are interconnected to form a region, and the various regions are
interconnected to form the core of the WAN.
The area could be
based on the number of locations to be connected with an upper limit of between
30 and 50. The area would have a star topology,
with the hubs of the stars linked to form the region. Regions could be geographic, connecting
between three and ten areas, and the hub of each region could be linked
point-to-point.
This three-layer
model follows the hierarchical design used in telephone systems. The links
connecting the various sites in an area that provide access to the enterprise
network are called the access links or access layer of the WAN. Traffic between
areas is distributed by the distribution links, and is moved onto the core
links for transfer to other regions, when necessary.
This hierarchy is
often useful when the network traffic mirrors the enterprise branch structure
and is divided into regions, areas, and branches. It is also useful when there
is a central service to which all branches must have access, but traffic levels
are insufficient to justify direct connection of a branch to the service.
The LAN at the
center of the area may have servers providing area-based as well as local
service. Depending on the traffic volumes and types, the access connections may
be dial up, leased, or Frame Relay. Frame Relay facilitates some meshing for
redundancy without requiring additional physical connections. Distribution
links could be Frame Relay or ATM, and the network core could be ATM or leased
line.
33.3 WAN Design
33.3.5 Other layered design models
Many networks do
not require the complexity of a full three-layer hierarchy. Simpler hierarchies may be used.
An enterprise
with several relatively small branches that require minimal inter-branch
traffic may choose a one-layer design. Historically this has not been popular
because of the length of the leased lines. Frame Relay, where charges are not
distance related, is now making this a feasible design solution.
If there is a
need for some geographical concentration, a two-level design is appropriate.
This produces a "star of stars" pattern. Again, the pattern chosen
based on leased line technology will be considerably different from the pattern
based on Frame Relay technology.
When planning simpler networks, the three-layer model
should still be considered as it may provide for better network scalability.
The hub at the center of a two-layer model is also a core, but with no other
core routers connected to it. Likewise, in a single-layer solution the area hub
serves as the regional hub and the core hub. This allows easy and rapid future
growth as the basic design can be replicated to add new service areas.
33.3
WAN Design
33.3.6 Other WAN design considerations
Many enterprise
WANs will have connections to the Internet. This poses security problems but
also provides an alternative for inter-branch traffic.
Part of the
traffic that must be considered during design is going to or coming from the
Internet. Since the Internet probably exists everywhere that the enterprise has
LANs, there are two principal ways that this traffic can be carried. Each LAN
can have a connection to its local ISP, or there can be a single connection
from one of the core routers to an ISP. The advantage of the first method is
that traffic is carried on the Internet rather than on the enterprise network,
possibly leading to smaller WAN links. The disadvantage of permitting multiple
links, is that the whole enterprise WAN is open to Internet-based attacks. It
is also difficult to monitor and secure the many connection points. A single
connection point is more easily monitored and secured, even though the
enterprise WAN will be carrying some traffic that would otherwise have been
carried on the Internet.
If each LAN in
the enterprise has a separate Internet connection, a further possibility is
opened for the enterprise WAN. Where traffic volumes are relatively small, the
Internet can be used as the enterprise WAN with all inter-branch traffic
traversing the Internet. Securing the
various LANs will be an issue, but the saving in WAN connections may pay for
the security.
Servers should be
placed closest to the locations that will access them most often. Replication
of servers, with arrangement for off-peak inter-server updates, will reduce the
required link capacity. Location of Internet-accessible services will depend on
the nature of the service, anticipated traffic, and security issues. This is a
specialized design topic beyond the scope of this curriculum.
Summary
An understanding
of the following key points should have been achieved:
- Differences in the geographic
areas served between WANs and LANs
- Similarities in the OSI model
layers involved between WANs and LANs
- Familiarity with WAN
terminology describing equipment, such as CPE, CO, local loop, DTE, DCE,
CSU/DSU, and TA
- Familiarity with WAN
terminology describing services and standards, such as ISDN, Frame Relay,
ATM, T1, HDLC, PPP, POST, BRI, PRI, X.25, and DSL
- Differences between
packet-switched and circuit-switched networks
- Differences and similarities
between current WAN technologies, including analog dialup, ISDN, leased
line, X.25, Frame Relay, and ATM services
- Advantages and drawbacks of DSL
and cable modem services
- Ownership and cost associated
with WAN data links
- Capacity requirements and
transit times for various WAN traffic types, such as voice, data, and
video
- Familiarity with WAN topologies,
such as point-to-point, star, and meshed
- Elements of WAN design,
including upgrading, extending, modifying an existing WAN, and
recommending a WAN service to an organization based on its needs
- Advantages offered with a
three-layer hierarchical WAN design
- Alternatives for interbranch
WAN traffic
No comments:
Post a Comment