Computer Networking: Network Design & Architecture

Running Head: Network Design & Architecture Proposal





















Computer Networking:
Network Design & Architecture


February 12, 2014

2

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Contents


Project Outline .................................................................................................................... 3
Distributed Network Requirements Analysis ..................................................................... 5
Communication Protocols Analysis and Recommendations .............................................. 7
The OSI Model ................................................................................................................ 8
The TCP/IP Model ........................................................................................................ 10
Advantages, Disadvantages, and Recommendations .................................................... 13
Network Traffic Analysis and Recommendations ............................................................ 15
Bandwidth and Consumption ........................................................................................ 17
Network Congestion and Control Algorithms .............................................................. 18
Network Design and Architecture .................................................................................... 20
The Infrastructure .......................................................................................................... 20
Software ........................................................................................................................ 23
Cloud Computing .......................................................................................................... 23
Future Needs Analysis and Recommendations ................................................................. 25
References ......................................................................................................................... 26


3

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Project Outline


In 2006 I bought a domain name, in hopes of starting a web development business to
create and host websites for individuals and small businesses. Time went on and career changes
left me with little to no time to work on creating a solid structure for this business. However, in
hopes of stepping foot in the right direction, I‟d like to research, develop and regain the
knowledge needed to make this dream a reality. I plan for this business to expand to have servers
in several locations to hold the information my customers will need to be competitive and
productive. The distribution of systems will create a better means of communication without the
hassle of excessive travel, and the ability to troubleshoot and teleconference with clients and
employees will allow me to keep costs low.
Though this organization is currently just a hypothetical business, I project it has the
potential to expand and reach a global customer base. I want to start off a little grounded with
very few clientele and later work to expand with employees in California, New York, Texas,
Washington, and of course here in Kansas. I want to see this business really take off within the
next two to five years, depending of course on the costs and allowable time to build and maintain
it. Therefore, through this project, I intend to gather the necessary information about distributed
networks, communication protocols, network traffic, network design and architecture, and the
possible requirements to maintain a strong, competitive edge for the future of my company.
Distributed networking will allow the company to expand globally by letting our users
access the information they need from their own locations as I will explain in the next section.
However, in building the appropriate network, I must examine other required information such
as communication protocols. So that my customers have a speedy and pleasant experience
communicating with my system(s), I must also examine the appropriate requirements for
network traffic. I would not want unhappy customers due to them not being able to access the
information they need in an efficient and rather effective manner. Next I will need to have a
network design in mind. This network structuring, or architecture, will allow me to create a fully
functioning network with the potential to span the globe. With this research and development of
4

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

information obtained, I will have a great start to building the web development business of the
future.
5

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Distributed Network Requirements Analysis


Distributed networks are, in its most general terms, a group of computers connected
together over many networks providing a single line of communication to retrieve data. The use
of a distributed network allows the processing of data to be easily accessible for many client
systems and allows for processing of data to be spread throughout the network. A network in and
of itself is a group of computing devices: computers, printers, servers, etc. A distributed network
allows many of these components to be connected over the internet to process vast amounts of
data and allows the processing of that data to occur at many different locations, not just on the
machine you happen to be working on at the time (Techopedia, 2010).
Distributed computing systems are a collection of independent computers that appear to
its users to be a single system. A couple of examples of distributed systems are the internet and
intranets. The internet of course is the net of nets where everyone has global access to data, and
services with no single authority. It is open ended meaning that it is so enormous that one could
spend a lifetime scanning the internet and never come to the end of its boundaries. An intranet is
a collection of computing devices like computers, printers, routers, switches, and servers that can
communicate with each other on a smaller scale without having to get out onto the internet to
process their data. Local area networks or LANs are a great example of distributed computing
systems via an intranet (Techopedia, 2010).
The most recent example of distributed systems is the mobility and ubiquitous computing
system. By embedding microprocessors in everyday devices, we now have refrigerators that
create shopping lists for us, watches that give us up to the minute weather reports, and of course
phones that can gather on location information for events that are happening in our specific area.
These systems are becoming ever more popular due to their convenience, popularity, and ease of
use. We are a society of instant gratification, therefore, the wave of ubiquitous computing has
exploded throughout the globe (Search, 2010). I believe this will be the most challenging part of
my research due the current lack of experience associated with mobile programming and mobile
6

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

accessibility. With our society being one of instant gratification, my customers will gain an even
greater projected client base because of the use of this technology. I myself have experienced
being somewhere I have not been accustomed to being and had to search out on my mobile smart
phone directions to local restaurants, places to buy goods, and the less expensive places for
services. Most of us have a phone in hand on a daily basis and could not even tell someone
where the nearest Yellow Pages is located. Better even, web pages not only provide information
to the customers on the name, address and contact information, they provide daily specials,
useful tips and tricks, embedded map generators that provide turn-by-turn directions on how to
get to their location. It just makes sense to be on the web.
The importance of distributed networks is to allow more clients the access to data quickly
and more effectively. We also have the ability to stretch out to other companies, other cities,
states and countries to obtain the information we need. In utilizing distributed systems and
distributed networks, businesses can expand their client base, companies can process more
information and the amount of data to be passed back and forth between companies and
consumers is no longer limited to catalogs, order forms, and snail mail.
Some of the challenges of creating or establishing a distributed system and distributed
networking is the age and ability of some of its pieces. Some networks within the system may
not be protected or even compatible with other pieces. Some may even be so old that the
information needing to pass will not be retrieved due to the complexity of its request. Updating
systems to accommodate such traffic within the system could be costly and unrealistic to
achieve. However, the benefits potentially outweigh the challenges based on the vast amount of
new consumers companies would gain from the upgrades (Usenix, 1999).




7

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Communication Protocols Analysis and Recommendations


In order to provide an analysis or a recommendation on communication protocols, I must
first gather information on what a communication protocol actually is. In doing so, I have taken
to the web, our book, and previous books I have examined in the past to find the required
information. In the simplest terms, a protocol is the way in which data is communicated back and
forth between systems. A protocol is defined in Tanenbaum as a set of rules governing the
format and meaning of the packets, or messages that are exchanged by the peer entities within a
layer (2011), simply put it is a set of rules used by communication devices, such as hardware
and/or software procedures, that allows communications to take place within a computer or
through a network (White, 2004). Information must meet certain criteria to be passed to and from
machines. Without meeting this criteria the information cannot be passed. Most systems will not
simply communicate directly to and from a single machine, the information will pass through
what‟s called a protocol stack. This is a list of protocols used by a certain system, one protocol
per layer (Tanenbaum, 2011), therefore, standards have been developed over the years to handle
the communication of information from machine to machine, or system to system.
These standards were developed to provide the appropriate rules and regulations for
information to pass or communicate, more so with all the different and complicated systems that
are so widespread and located globally, these standards play a huge part in allowing the
communication of information to process error free and without complication or congestion,
which I will explain later in this project. Network architecture is a communications model that
sets up layers for information to pass through. It consists of the philosophy and concept for
enabling communications between multiple locations and multiple systems. The goal of network
architecture is to promote an open, simple, flexible and efficient telecommunications
environment by using, among other standard equipment and interfaces, standard protocols
(Blume, 2010). When a command is sent through the system, it will pass through several layers
before it is actually sent out to another system to complete the communication loop. Two very
popular architectures that are currently in use are the OSI model and TCP/IP model. These two
8

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

architectures allow for message passing by adding routing information as headers, piecing
together packets of information so that the lines of data are easily sent and received, and by
making sure that the information is passed cleanly and effectively.

The OSI Model

The most basic standard in practice since the early to middle „70s has been the OSI
model. Developed by Honeywell Information Systems, headed by Mike Canepa and Charlie
Bachman for the purposes of distributed database development, the original seven layer
infrastructure was the 1977 result of collaborated work being done for IBM‟s system network
architecture (SNA), protocols for ARPANET, and concepts of presentation services in
development for standardized database systems; they had internally referred to it as distributed
systems architecture or DSA. In 1977, the British Standards Institute had proposed to the ISO
(International Organization of Standards) that suggested an international standard needed to be
developed for distributed systems. At that point is when the American National Standards
Institute (ANSI) was put to task to develop proposals to satisfy this need. Bachman and Canepa
joined in the development meetings held by the ISO. They provided their model and without
contest, after a bit of fine tuning, it was put into practice in June of 1979 (Stallings, 1998)
The current standard OSI (Open System Interconnection) model cannot be described as a
single process, but a layer of interconnected processing layers that determines how a
communication should and will be sent. The OSI model is the most commonly used model by
most all major computer and network vendors. It is comprised of seven layers being the physical
layer, the data-link layer, the network layer, the transport layer, the session layer, the
presentation layer, and the application layer.
 The physical layer deals with the raw bits of data getting from point A to point B no matter
what condition they are in over the mechanical and electrical portion of the network medium,
the cable.
9

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

 The data-link layer handles putting together, or assembling, the data into single units or
blocks to allow for flow and error control. It is typically divided into two parts depending on
the need and use, Logical Link Control (LLC) and Medium Access Control (MAC).
 The network layer establishes, maintains and terminates logical and/or physical connections
and is responsible for translating logical addresses into physical addresses. It also provides
network routing and flow control functions across the interface.
 The transport layer makes sure that the data is successfully sent and received between the
two systems. If a transmission errors out, it has the responsibility to request the data be
resent.
 The session layer decides when to end communication between systems. It also provides the
mechanisms that control data exchange and coordinates the interaction. The session layer is
also responsible for setting up and ending communication channels between two
communicating components.
 The presentation layer performs all the code conversions and data reformatting so that the
data is in the correct form to be received by the receiving application.
 The application layer provides the user interface between the software running in the
computer and the network. It also provides functions to the user‟s software, including file
transfer access and management (FTAM) and electronic mail.














10

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______









Figure 1 below demonstrates the process of sending a simple message through the layers.



Figure 1: Networked computers communicating through the OSI model (Novell, 2013)


The TCP/IP Model

11

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Originally developed by the United States Defense Advanced Research Projects Agency
(DARPA or ARPA) in 1973, the TCP (transmission control program) model began its lifecycle
not as a means to process distributed information over the internet, but as a fully encompassing
protocol. However, it was discovered by Jon Postel in 1977 memo showing that this model was
performing too much and would eventually become a nonfunctional entity. Postel‟s memo stated
that TCP needed a few modification to perform swift and smooth functionality for the purposes
of transmitting information across the internet. He believed that the, at that time, current design
of internet protocols was violating the principles of layer by performing the host level end to end
protocol functions and by serving as an internet packaging and routing protocol. Postel suggested
dividing the functions of the TCP into a layered model following the example of the OSI model,
where TCP handles layer three activities and IP handles layer four activities of the OSI model for
use in communications across the internet. Which brings us to the current TCP/IP (Transmission
Control Protocol/Internet Protocol), version 4 developed in 1980 (Kozierok, 2005).
The TCP/IP model is interconnected with the OSI model in that a communication being
sent or received via the internet will have to process through these protocols at both ends of the
communication. As data moves through the layers, header information is added or removed for
compatibility purposes between systems over the network. The four layers of the TCP/IP model
are the network interface layer, the internet layer, the transport layer, and the application layer.
 The network interface layer is the lowest layer and is responsible for putting frames on
the wire and pulling frames off the wire. It is equivalent to the data-link layer of the OSI
model.
 The internet layer provides three services – a connectionless delivery service, a
mechanism to break data into individual packets or frames on the transmitting side and
put them back together on the receiving side, and the routing functions necessary to
interoperate with other networks. It is equivalent to the network layer of the OSI model.
 The transport layer provides communication sessions between connected computers and
is responsible for error detection and correction. It is equivalent to the transport layer of
the OSI model.
12

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

 The application layer serves as a window for users and application processes to access
network services. It is equivalent to the application, presentation, and session layers of
the OSI model.

Figure 2 below shows the layers for both the TCP/IP model and the OSI model and how
they relate to each other.


Figure 2: TCP/IP model and OSI model side by side (DreamCloud, 2013)

Figure 3 shows each of the TCP/IP model layers in reference to the layout of the OSI
model. You can also see the protocols associated with each layer in reference to the transmission
of information over the internet.
13

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______


Figure 3: The TCP/IP Suite

Advantages, Disadvantages, and Recommendations

The main advantages of using TCP/IP across a distributed network is due to its ability to
provide compatibility with the internet. It allows for communication across interconnected
networks with different operating systems and hardware architectures. TCP/IP also provides for
routing support. Further advantages are that it is an industry standard and is not owned by one
single company or provider. TCP/IP has a scalable client/server architecture and it enables inter-
networking between organizations. However, some of the disadvantages of TCP/IP are that it can
be intricate to set up and manage, and difficult to administer, it can be slower than its newer
14

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

counterparts in smaller LANs, and depending on the networking devices available, it can be a bit
costly in some network designs.
TCP/IP has been the backbone for internet transmission of data for decades and has been
manipulated in many cases for improvement from its beginnings at versions one, two and three.
Version four has been successfully assisting organizations, companies and individuals transmit
information over the internet for years, and with the newest format of version six opens up all
new possibilities for transmission control and information processing. I believe that the company
could greatly benefit from setting up a network platform structured around the concepts of the
OSI model and TCP/IP as a basis.




15

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Network Traffic Analysis and Recommendations


Blume defines network congestion as “a condition when a communications link, path or
network, experiences a load that exceeds its capacity.” Two most common congestion control
methods in distributed networks is the use of buffering and the discarding of data/packets. In the
more localized environment of a LAN, network segmentation is the preferred method by using
routers, switches and bridges to create more direct pathways for data to flow. This eliminates the
cause of the congestion from the get go. By doing so, larger more complex networks can be
divided up into smaller networks (Blume, 2010).
In order to better understand congestion in general we must first look at how these
devices that aide in prevention actually function. A router is an inter-network device that
connects networks together and determines how and when packets or messages are received. It is
also responsible for determining if those packets are meant for its LAN and if so, sends the
packets to the appropriate device. If those packets are not meant for its LAN, they are discarded,
keeping unnecessary traffic off of the LAN. A switch is a device that performs the multiple
functions of filtering, flooding, and transmitting of frames. Switches are not used to create inter-
connected networks, but are instead used to make the processing and functioning of the networks
better, faster, more efficient. Switches do not and cannot forward IP packets to other networks,
they can only switch frames between its ports. By default, each port of a switch is its own
individual collision domain. A bridge basically operates much like a switch, however, design
limitations make a switch much more effective for larger networks (Blume, 2010). Figure 4
below provides and visual explanation of the network routing and switching design.

16

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______


Figure 4: A visual example of network routing and switching (Adtran, 2013)

The use of routers and switches to aide in the elimination of network congestion has its
advantages and disadvantages. For example the advantages would include the ability to break up
larger networks and section the information out to direct feeds upon where the information needs
to go. Also, packets of information that are not intended to be processed on that network are
dropped so to eliminate unnecessary traffic on the network. That being said, these advantages
also create disadvantages. Logically speaking, I see dropped packets on faulty systems in my
current line of work all the time. It creates frustration for my customers. Usually has to do with
the quality of service and the quality of hardware feeding the information to the customer‟s
network. However, in the case of routers and switches, the packets received by the equipment is
just received, sent, and passed along to where the packets tell it that it needs to go. If the coding
of the packet being directed does not have the correct information to determine an appropriate
location to be sent to, it is dropped and lost. This does not make for a very effective or efficient
means of processing information. Another disadvantage of using multiple routers and switches to
17

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

divide larger networks into smaller networks is the cost in doing so. Routers and switches,
especially multiport, multiprocessing devices capable of moving information a record speeds is
very costly, depending on the amount of information a company needs to process. For my
business, trying to price out an effective and efficient model for any small business struggling in
the current economy would definitely be a hard sell. Therefore, it may be in my best interest to
determine exactly what types of information I will processing for myself and for clients, and
determine the potential causes of congestion on the networks I intend to create.

Bandwidth and Consumption

Network traffic and bandwidth consumption plays a huge role in the ease of operation for
businesses and consumers alike. I have experience in addressing customer complaints of slow
internet browsing. The best solution can be the least complicated fix, however, in dealing with
large networks, the simplest fix is not always the best solution. To determine the best solution I
will need to examine several key aspects of network traffic and address the major uses of the
distributed network one level at a time. It will be important to determine the requirements of my
distributed network based on the tasks that will be performed routinely. Video conferencing,
voice streaming, data transmission, and web hosting will be the four bandwidth consumers the
business will have to monitor for congestion. With this being a distributed network, the problem
of data being bottlenecked will be a greater possibility due to the amount of users trying to
access the information at the same time.
In order to communicate over the network via video conferencing, I will need adequate
bandwidth, traffic support in a dedicated class, call admission control for bandwidth
management, a multipoint conferencing unit or bridge, and management and monitoring of the
WAN vendor. I know that video conferencing sessions can consume extremely large amounts of
bandwidth, in some cases as much as 6 Mbps per call. In order to find the adequate amount of
bandwidth for video conferencing, determine the bandwidth of each call by comparing resolution
and frame rates over the settings that work best for the company. Next, determine the maximum
18

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

number of concurrent calls being conducted simultaneously. Multiply the bandwidth by the
number of calls to get the required bandwidth for each network access link, then multiply that
number by the amount of links that are intended and that number is the required amount of
bandwidth needed for video conferencing.
Network Congestion and Control Algorithms

Blume defines network congestion as “a condition when a communications link, path or
network experiences a load that exceeds its capacity.” Two most common congestion control
methods in distributed networks is the use of buffering and the discarding of data/packets. In the
more localized environment of a LAN, network segmentation is the preferred method by using
routers, switches, and bridges to create more direct pathways for data to flow. This eliminates the
cause of the congestion from the get go. By doing so, larger more complex networks can be
divided up into smaller networks (Blume, 2010).
In order to better understand congestion in general we must first look at how these
devices that aide in prevention actually function. A router is an inter-network device that
connects networks together and determines how and when packets or messages are received. It is
also responsible for determining if those packets are meant for its LAN and if so, sends the
packets to the appropriate device. If those packets are not meant for its LAN, they are discarded,
keeping unnecessary traffic off of the LAN. A switch is a device that performs the multiple
functions of filtering, flooding, and transmitting of frames. Switches are not used to create inter-
connected networks, but are instead used to make the processing and functioning of the networks
better, faster, more efficient. Switches do not and cannot forward IP packets to other networks,
they can only switch frames between its ports. By default, each port of a switch is its own
individual collision domain. A bridge basically operates much like a switch, however, design
limitations make a switch much more effective for larger networks (Blume, 2010).
The use of routers and switches to aide in the elimination of network congestion has its
advantages and disadvantages. For example the advantages would include the ability to break up
larger networks and section the information out to direct feeds upon where the information needs
19

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

to go. Also, packets of information that are not intended to be processed on that network are
dropped so to eliminate unnecessary traffic on the network. That being said, these advantages
also create disadvantages. Logically speaking, I see dropped packets on faulty systems in my
current line of work all the time. It creates frustration for my customers. Usually has to do with
the quality of service and the quality of hardware feeding the information to the customer‟s
network. However, in the case of routers and switches, the packets received by the equipment is
just received, sent, and passed along to where the packets tells it it needs to go. If the coding of
the packet being directed does not have the correct information to determine an appropriate
location to be sent to, it is dropped and lost. This does not make for a very effective or efficient
means of processing information. Another disadvantage of using multiple routers and switches to
divide larger networks into smaller networks is the cost in doing so. Routers and switches,
especially multiport, multiprocessing devices capable of moving information a record speeds is
very costly, depending on the amount of information a company needs to process. For my
business, trying to price out an effective and efficient model for any small business struggling in
the current economy would definitely be a hard sell. Therefore, I need to determine other
effective avenues to keep congestion to a minimum to non-existent.
End-to-end congestion control requires the sender and receiver to provide some sort of
feedback information telling the sender that there is congestion and to slow down. In order to
have an efficient throughput of traffic, we cannot simply rely on the Transport Control Protocol
(TCP). The information gathered at just the end nodes of the system is not sufficient enough to
be effective in larger networks. Therefore, a need for a control mechanism like AQM, Active
Queue Management. These enlist the help of a single Explicit Congestion Notification (ECN) bit
placed in the packet header. This bit provides instant feedback to internet gateways about
potential congestion to the end users, or end nodes. Some additional processes for congestion
control are known as congestion control algorithms (CCAs). These algorithms include traffic-
aware routing, admission control, load shedding, and random early detection algorithms (RED).
Traffic-aware routing allows networks with peak times of use gain extra bandwidth from areas
with lower usage during non-peak times. If the packets being sent will cause the network to
20

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

become congested, the use of admission control with block these packets from causing the
congestion, and if there is still too much traffic on a network and the congestion has yet to be
eliminated, load shedding can be used to dump packets and drop them from the network before
the limit of buffer space has been depleted. Similar to load shedding is random early detection,
similar in the sense that packets are dropped from a network transmission that will cause
congestion. Their difference, however, is that RED picks packets at random and sends a signal
back to the fastest senders for them to see the dropped packets (Tanenbaum, 2011).

Network Design and Architecture

In determining the appropriate setup, layout and overall design of the needed network
architecture for N2IT4U, recommendations for communications and network traffic must be
established. I will need to draw up a model of the network, provide a description, and give
substantial reasoning behind my decision. I will also need to define the necessary hardware and
software for this new design. Now because this is my business, I will need to make sure all
elements of this design are well within my budget. I will also need to determine an appropriate
time frame for implementation. With my goal being set for the next two to five years, I will have
a little bit of time to research the best and most cost effective hardware such as servers,
gateways, router, switches, and the newest development in network cabling, and the use of
mobile equipment. I also know that I will need to negotiate cost effective solutions for multiple
providers and create a little redundancy for the purposes of backup and network safeguarding.
My network should be fully functional 365 days per year regardless of service outages or carrier
issues.

The Infrastructure

21

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

In order to design a network architecture that will be adequate for N2IT4U, it must be
able to transfer the appropriate amount of data with the least amount of congestion in the most
cost effective approach. I spoke earlier about the use of routers and switches, but did not touch
on mobility at that time. Mobile access to processes and information would be a valuable
advantage for N2IT4U.com. The ability to be mobile would allow our customers, clients and
employees gain the information that they need to function efficiently and effectively. With the
use of tablet devices rather than bulky stationary pcs, the ability to be mobile is a definite reality.
Our infrastructure will need to also allow for the applications on those tablets to store and
retrieve information quickly and efficiently at any given time. We use a similar process where I
currently work and with the use of iOS devices we are able to test, gather information, process
requests and function almost as effectively as we did when we were still using laptop computers
-- Toughbooks. I say almost because the iOS devices are limited and cannot be expanded in any
way to allow for the use of the same external devices we had available to use with rugged
computing devices. Another current downfall of these particular mobile devices is that they
require a service carrier such as AT&T, Verizon, Sprint, etc. to carry the data over cellular
towers to process the data. Mobile hot spots are not always readily available in more rural areas
and can create functionality issues with this particular type of equipment. However, the
correction of cellular imperfections is for a later time and discussion. For the purposes of this
proposal, we will assume cellular data is near perfect and can be accessed anywhere.
The figure below demonstrates a visual interpretation of a distributed network with
mobile elements. Each one of the little “disconnected” pods could represent a mobile hot spot or
a cellular tower. The larger blocks here represent the local LANs that branch out to other centers
or areas. This is a very good interpretation of the layout I would like to see my company grow to.
The larger blocks being the main hubs in Kansas, Texas, Washington, California, New York and
anywhere else we decide to branch out to. The smaller “connected” blocks are the
communication locations such as offices and server locations that house the information and
communications relays. The “non-connected” smaller blocks represent the mobile network of
data transfer capability I would like to see put in place for our mobile clients and customers. That
22

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

all being said, this hardware version of the infrastructure needs the ability to process information
cleanly and effectively. Therefore, adequate security and software features must be put into place
to keep our information and hardware safe and secure.



Figure 5: Visual Example of a distributed network infrastructure with mobile nodes

In breaking down the costs and for the needed hardware, I took to the internet to browse
leading companies dealing with mass business networking equipment. Depending on the
capacity of switching and controlling to be done by the hardware specifically, the cost of
equipment ranges between around $500 to $25,000 each. I would like to have a little control over
MAC address handling and some hands on control and functionality. Some systems have this
control built in to be automatically controlled by the device, but those pieces of equipment are
well into the higher price ranges nearing $50 to $60,000. Maybe later on down the road, but for
right now, my eye is on the middle of the road with some functionality in cloud devices as well.
Cloud services are a lot more affordable, but have risky potentials that I may not be willing to
put all of my trust into at this point.

23

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Software

There are new companies out there like SalesForce.com, Workday, and NetSuite just to
name a few (Bort, 2013). These companies are becoming the wave of the future with their
software-as-a-service (SaaS) mantra. Salesforce.com charges $125 per month and billed annually
for its business and professional services and was the only company I found that publicly
advertised its pricing online. In starting out, the software requirements for this elaborate system
of distributed locations I plan to venture into might be better suited for portions of use in the
form of SaaS, especially for such a diverse range of customers and employees. We need as much
functionality as possible with the most mobile and accessible means of communication. The use
of cloud computing can lead us in the right direction without burning a gaping hole in my
pocketbook.
In searching for the best means to deliver my clients and employees the ability to work
with our information, I would also like to look towards mobility and cloud computing. These can
pose a great advantage for the expansion of the company to grow and be successful long into the
future.




Cloud Computing

Cloud computing is defined as a general term for anything that involves delivering hosted
services over the internet (TechTarget, 2010). As a whole, cloud computing IS distributed
computing in that all the consumer needs to access the information is a connection to the internet
and the device they so choose to access it with. Cloud computing can either be public or private
where public cloud provided services are sold to anyone on the internet and private services are
sold over a proprietary network or a data center that supplies hosted services to a very limited
number of people. These public clouds can also be purchased by businesses to create their own
24

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

virtual private clouds. Either way, the intended goal of these cloud services is to provide
customers with easy access to resources they may otherwise not be able to obtain through
networking alone (TechTarget, 2010).
Forms of services provided via cloud computing are Infrastructure-as-a-Service (IssS),
Platform-as-a-Service (PaaS), and Software-as-a-Service (SaaS) as mentioned previously. IaaS,
often referred to as Utility Computing because of its on demand usage capability, is a means for
businesses to start, stop, access and configure virtual servers and storage on a “use as you go”
basis. PaaS is the tools necessary for businesses to provide software and product development
tools to their customers by hosing these products over their cloud infrastructure, however,
currently the software developed on this type of system is proprietary and cannot be moved from
the provider‟s platform. So if it‟s built here, it must stay here, this is very good information to be
aware of. This could be very important if I choose to either host a cloud with PaaS, or subscribe
to PaaS via another provider‟s system. SaaS is the most mobile of these defined services. The
provider supplies both the hardware and the software because these services range anywhere
from Web-based email to inventory control and database processing. The service provider hosts
the application and the data leaving the customer to the freedom of mobility as long as the device
they are using is capable of accessing the hardware and software the provider is offering
(TechTarget, 2010).




25

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Future Needs Analysis and Recommendations

The future of networking and computing is ever changing, and it is very rapid. The most
current of information shows that the advances in mobility and cloud computing are definitely
the future of pristine business flow and practice. In order for N2IT4U to be successful now and
many years into the future, it would be a wise move for me to make proper use of this cloud
technology, in IaaS and SaaS. What would be even more reasonable is to have a strong foothold
on hosting these technologies so that the costs outweigh the means. So for the business practices
of N2IT4U.com, I would strongly recommend looking into the use of servers capable of hosting
such services and the data I intend to provide to my customers on an on demand basis. I will also
need routers and other hardware and devices for my employees to process and maintain the
information I want to gather and process. That being said, with the I believe N2IT4U.com has
the potential to set foot in the right direction as being a leader in communications and
development, now and for years to come.


26

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

References


Adtran (2013) VLAN Switching and Inter-VLAN Routing. Retrieved December 8, 2013 from the
World Wide Web: http://www.adtran.com/web/page/portal/Adtran/group/2870

Blume, Steve (2010) Data Communications & Networks. Carlsbad, CA: Applied Professional
Training. (Original work published 1994)

Bort, Julie (2013) The 15 Most Valuable Cloud Computing Companies In The World Are Worth
Way More Than You‟d Think. Business Insider. Retrieved Decemeber 12, 2013 from the
World Wide Web: http://www.businessinsider.com/the-15-most-valuable-cloud-
computing-companies-2013-7?op=1

DreamCloud (2013) IP Basic. Retrieved December 5, 2013 from the World Wide Web:
http://dreamcloud.artark.ca/ip-basic/

Kozierok, Charles M. (2005) “NFS Architecture and Components” The TCP/IP Guide. Retrieved
December 5, 2013 from the World Wide Web: http://www.tcpipguide.com/free/
t_NFSArchitectureand Components.htm

Novell (2013) Novell’s Networking Primer: Network Communications through the OSI Model.
Retrieved December 5, 2013 from the World Wide Web: http://www.novell.com/info/
primer/prim05.html

Search Networking (2010) Pervasive Computing (Ubiquitous Computing). Retrieved November
24, 2013 from the World Wide Web: http://searchnetworking.techtarget.com/definition/
pervasive-computing

Stallings, William (1998) The Origins of OSI. Retrieved December 5, 2013 from the World Wide
Web: http://williamstallings.com/Extras/OSI.html

Tanenbaum, A.S. & Wetherall, D.J. (2011) Computer Networks (5
th
Ed). Boston, Massachusetts:
Pearson Education, Inc., publishing as Prentice Hall.

Tecnick (2011) Network Switching Tutorial. Retrieved December 5, 2013 from the World Wide
Web: http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=guide_networking_
switching

27

Network Design & Architecture Proposal
©2014 N2IT4U™ Wichita, KS All rights reserved

(Document Source Information: Heffron-Blake, Darla. (2014, February). Computer Networking:
Network Design & Architecture. Wichita, KS: N2IT4U, Retrieved #DATE#, From http://______

Techopedia (2010) Distributed Network. Retrieved November 24, 2013 from the World Wide
Web: http:// www.techopedia.com/definition/27788/distributed-network
TechTarget (2010) Cloud Computing Retrieved January 7, 2014 from the World Wide Web:
http://searchcloudcomputing.techtarget.com/definition/cloud-computing

Usenix (1999) Massively Distributed Systems: Design Issues and Challenges. Retrieved
November 24, 2013 from the World Wide Web: https://www.usenix.org/legacy/events/
es99/full_papers/ nessett/nessett.pdf
White, Curt M. (2002) Data Communications & Computer Networks (2
nd
Ed). Boston,
Massachusetts: Thomson Learning, Inc., publishing as Course Technology.

White, Curt M. (2004) Data Communications & Computer Networks (3
rd
Ed). Boston,
Massachusetts: Thomson Learning, Inc., publishing as Course Technology.