NETWORK DESIGN

D16-B
GROUP-1
 SYLLABUS

Module 4: Network Design (8 hrs)

• 4.1
● Network layer design, Access layer
design,access network capacity,network
topology and Hardware, completing the
access network design
 PREVIOUS YEAR QUESTIONS
Dec 2017
● Explain in detail, access layer design (10
Marks)
Dec 2016
• What do you mean by Network layer
design? (5 Marks)
May 2016
• Describe Ubiquitous Access and
Hierarchical access and compare them
(10 Marks)
• Write a note on steps for completing
Access network design (5 Marks)
 IMPORTANT QUESTIONS
1.What are the parameters to be considered for
Network Designing?
2. Explain the process to access network
capacity?
3. Differentiate between switching and routing?
4. Compare and contrast IPv4 and IPv6 header
fields?
5. Explain Collapsed Backbone Architecture?
6. Explain Bridged Backbone and Routed
Backbone architecture?
 NETWORK DESIGN LAYERS
• The four general layers of network
design are as follows:

• Application
• Premises
• Access
• Backbone
NETWORK DESIGN LAYERS
• CORE: This is the core or backbone of the
network. The core needs to be reliable and
fast. It typically has fault tolerance as a
design goal.

• DISTRIBUTION: This layer is about

control. It implements network policies for
security, routing control and filtering.

• ACCESS: Collision domain segmentation

accomplished with Ethernet switches falls
in this layer.
CISCO’S THREE LAYER
HIERARCHIAL MODEL
 ACCESS LAYER DESIGN

• Access Layer Design involves selecting or

modifying LAN, MAN, or dial dedicated
across topology.
• Before beginning the process of traffic
analysis, capacity planning and network
infrastructure design, we need to design
the access layer.
• We need to access information collected
in five key areas, as explained in the
further slides.
 PHYSICAL CONNECTIVITY

Is there any existing network? If so what are

the physical and logical configurations and
why?

● Sometimes the access network is designed

based on what is available verses what is
desired.
● The widest variance of any connectivity
requirements are the physical media level.
 PHYSICAL CONNECTIVITY

● Copper, fibre and air are the three most

common media for physical connectivity.

● The access layer design needs to

accommodate user-side premises interfaces
such as Ethernet, Token-Rings and ATM
LANs hosts.
 PHYSICAL CONNECTIVITY

● The interface from access layer design into

backbone layer is typically on a router or
switch using serial or telephony interfaces.

● The typical options of building wiring are

UTP, STP, thin and thick coax and glass or
plastic fibre optic media.
 PROTOCOLS
• Networks that have been around for a long
time are typically multiprotocol and requires
some consolidation for optimal design.

• Protocol support can range from providing

transparent transport for a homogeneous
protocol to multiple protocols.

• Each protocol that operates over each

interfaces must be defined by identifying
syntax, semantics, timing, implementations
and peculiarity.
 PROTOCOLS

• Determine which portion of protocol is used.

• There will be most likely be many dissimilar

protocols in the application layer and most
will not communicate with each other.

• As IP became the dominant networking

protocol, many vendors such as Novell joined
the bandwagon by supporting IP versions of
systems. Novell Netware 5 allows use of IP
or IPX to the user.
SWITCHING V/S ROUTING
 QUALITY OF SERVICE (QOS)

• Which of your applications need support of

QOS and what level of QOS? Will class of
service (COS) do?

• Being able to prioritize traffic in both LAN

and WAN is becoming more important every
year as voice and video traffic migrate to
data network.
 QUALITY OF SERVICE (QOS)

• Some access devices such as routers,

switches and multiplexers allow
prioritization of user traffic.

• Prioritization can be arranged through WAN

side of access device and can be performed at
packet level of IP services.
 QUALITY OF SERVICE (QOS)

• Ethernet switches can support various class

of service settings.

• The three main OS techniques are

 QUALITY OF SERVICE (QOS)

• IEEE 802.1 P/Q standard functioning at

Layer 2 OSI Reference Model.

• IP-Precedence TOS (Type of Service) in

IPv4.

• Vendor specific solutions that can tag

based on port or MAC address functioning
at Layer 1 and 2 of OSI Reference Model.
QUALITY OF SERVICE (QOS)
QUALITY OF SERVICE (QOS)
IPV4 & IPV6 HEADER CONTRAST
 ACCESS NETWORK CAPACITY

• Access Network Capacity is well known if

user applications and inputs are well
known.

• But it is extremely unpredictable due to:

• Bursty LAN traffic pattern.
• Peer-to-peer communication pattern.
• many-to-many communication
pattern.
of client server applications.
 NETWORK TOPOLOGY AND
HARDWARE
• Includes LAN , Remote access and dedicated
access methods.

• While selecting network topology following

options must be reviewed:
▪ Scalability and flexibility
▪ Reliability and availability
▪ Performance
▪ Security
 NETWORK TOPOLOGY AND
HARDWARE

• Two major Topologies:

• Ubiquitous Access
• Hierarchical Access
 UBIQUITOUS ACCESS

• The ubiquitous access approach allows all

users to transmit and receive data from all
areas of the network across a shared device
or medium.
 UBIQUITOUS ACCESS

• Any user or any LAN can pass data through

hub or routing device as well as access any
other device in the network.

• This style of access is good for organization

that are spread out and non-structured in
nature.
 UBIQUITOUS ACCESS
• This style provides one-dimensional flat
network and used with smaller network.

• Performances with performance degradation

are easily diagnosed and the network can be
more difficult to manage.

• It offers a very large single point of failure

that can take every user down at one time.
UBIQUITOUS ACCESS
 HIERARCHICAL ACCESS
• Hierarchical access provides a user access
hierarchy in which traffic defined for its own
local metropolitan and wide area remains in
that geographical area.
• Each user is connected to a floor LAN, each
LAN is connected to a building LANs are
connected to one campus MAN and the
campus MAN is connected to the WAN.
• Connectivity between segments is
accomplished via bridges, routers or LAN
switches.
 HIERARCHICAL ACCESS
• It is used in larger network.

• Hierarchical access topology allows the

capability of connectivity low speed LANs to
high speed LANs, MANs, WANs etc.

• Hierarchical style network helps to protect

users from broadcast storms and mark it
easier for administration to regulate and
secure traffic flows between segments.
HIERARCHICAL ACCESS
 UBIQUITOUS HIERARCHICAL
ACCESS ACCESS
1. Performance is low 1. Performance is high
due to enhanced
segments

2. No Protection from 2. Protection from

Broadcast storm Broadcast storm
3. Isolation and 3. Isolation and diagnosis
diagnosis of problems of problems are easier
are difficult and slow and fast

4. Less traffic 4. More traffic

5. No cost of network 5. Additional cost of
equipment the network
equipment and
services to localize the
network
6. Security 6.Security
administration is not administration is easy
easy to regulate and to regulate and secured
traffic flow between traffic flow between
segment is not secured. segment

7. They are spread out 7. They are used for

and non-structured in structured and layered
nature. It provides 1-D networks.
flat network and is
used with smaller
network.
 COMPLETING THE ACCESS
NETWORK DESIGN
The steps for completing access network design
include verifying application intelligence,
confirming design and choice of access devices,
choosing the number and type of access devices
and verifying total access topology.
 Application and Protocol
Intelligence Verification

➢ When the access network was performed it

was assumed that there was some level of
intelligence in the user devices and
applications.
➢ The greater this level of intelligence, the
less network intelligence that is required
from the access and backbone network and
more the access and backbone network can
concentrate on passing data.
➢ Confirm where that intelligence lies before
the application get to the access network.
 ACCESS DESIGN LEVEL

• The next level of verification is of access

device.
• The access device should be the feeder
portion of the network design, where user
applications, devices and protocols of a
defined geographic area are fed into a single
or multiple devices.
• The access device will then communicate
with other access devices or directly with the
backbone.
 ACCESS DESIGN LEVEL

• The typical access node will aggregate

transmit data from different interfaces to the
backbone via a single protocol.
• Access device should reduce number of access
ports, protocols and data formats fed to the
backbone.
• ATM is one technology that can offer
integrated access of all forms of
communication, data, voice and video.
 NUMBER AND TYPE OF ACCESS
NODE

• The number of access node is determined by

traffic matrix, optimized grouping of matrix,
optimized grouping of traffic and placement
options for access nodes.
• The network design should always allow for
a multiple node failure as well as alternate
routes for a minimum of a single failure
 NUMBER AND TYPE OF ACCESS
NODE

• No single site failure should take entire

network down.
• Most router and switch vendors make almost
all common protocols available for their
devices either at no cost or very low cost.
 BACKBONE NETWORK
ARCHITECTURES

• Identifies the way backbone interconnects LANs.

• Defines how it manages packets moving through

BB.
 BACKBONE NETWORK
ARCHITECTURES

• Fundamental architectures:
● Bridged Backbones
● Routed Backbones
● Collapsed Backbones
• Rack-based
• Chassis-based
● Virtual LANs
• Single-switch VLAN
• Multi-switch VLAN
BRIDGED BACKBONE
 BRIDGED BACKBONE
• Move packets between networks based on their
data link layer addresses.

• Cheaper (since bridges are cheaper than routers)

and easier to install (configure).
• Just one subnet to worry.
• Change in one part may affect the whole
network.
 BRIDGED BACKBONE

• Performs well for small networks.

■ For large networks broadcast
messages (e.g., address request,
printer shutting down) can lower
performance.

• Formerly common in the distribution layer.

• Declining due to performance problems.
ROUTED BACKBONE
 ROUTED BACKBONE

• Move packets using network layer addresses.

• Commonly used at the core layer.

• Connecting LANs in different buildings in the
campus.
• Can be used at the distribution layer as well.
• LANs can use different data link layer protocols.
 ROUTED BACKBONE

• Main advantage: LAN segmentation.

• Each message stays in one LAN; unless
addressed outside the LAN.
• Easier to manage.

• Main disadvantages.
• Tend to impose time delays compared to
bridging.
• Require more management than bridges &
switches.
COLLAPSED BACKBONE
 COLLAPSED BACKBONE

• Replaces the many routers or bridges of the

previous designs.
• Backbone has more cables, but fewer devices.
• No backbone cable used; switch is the backbone.
 COLLAPSED BACKBONE

• Advantages:
• Improved performance (200-600% higher).
• Simultaneous access; :switched”
operations.
• A simpler more easily managed network –
less devices.

• Two minor disadvantages:

• Use more and longer cables.
• If the central switch fails, the network
goes down.
THANK YOU