Network Design Edited
Network Design Edited
Your network design could take into consideration many technologies such as Token Ring,
FDDI, and Ethernet. This design will focus on the Ethernet technology as that is the technology
you will most likely encounter when you plan future designs. Ethernet has a logical bus
topology, which leads to collision domains; however, you will try to keep them small by using
the process called segmentation. Once you have settled on Ethernet, you must develop a
Layer 1 LAN topology. You must determine the type of cable, and the physical (wiring)
topology that you will use. The most common choice is CAT 5 UTP as the medium, and an
extended star topology as the physical (wiring) topology. Then you must decide on which one,
of the several types of Ethernet topologies, you need to use. Two common types of Ethernet
are 10BASE-T and 100BASE-TX (Fast Ethernet). If you have the resources, you might run
100BASE-TX throughout the network. If not, you might use Fast Ethernet to connect the main
distribution facility (central control point of our network) to other intermediate distribution
facilities. You might use hubs, repeaters, and transceivers in your design, along with other
Layer 1 components such as plugs, cable, jacks, and patch panels. To finish Layer 1 design,
you must generate both a logical and a physical topology. (Note: As always, an important part
of your design involves documenting your work.)
The next step is to develop a Layer 2 LAN topology, that is, to add Layer 2 devices to your
topology to improve its capabilities. You could add switches to reduce congestion and collision
domain size. In the future, you may be able to afford to replace hubs with switches, and other
less intelligent Layer 1 devices with more intelligent Layer 2 devices.
The next step, then, is to develop a Layer 3 topology; that is, to add Layer 3 devices that will
add to the topology's capabilities. Layer 3 is where routing is implemented. You could use
routers to build scalable internetworks such as LANs, WANs, or networks of networks. Routers
will impose logical structure on the network you are designing. They can also be used for
segmentation. Routers, unlike bridges, switches, and hubs, break up both collision and
broadcast domains.
The LAN's link to WANs and to the Internet must also be considered. As always, you should
document your network design's physical and logical topologies. Your documentation should
include any brainstormed ideas, problem-solving matrices, and any other notes you made
while making your determinations.
8.1.2 Network design issues
In order for a LAN to be effective and serve the needs of its users, it should be implemented
according to a systematic series of planned steps. While you are learning about the design
process, and creating your own designs, you should use your engineering journal extensively.
Your first step in the process is to gather information about the organization. This information
should include:
Hopefully, this step will also help you identify and define any issues or problems that need to
be addressed (e.g. you may find that a remote room in the building may not have network
access.)
The second step is to make a detailed analysis and assessment of the current and projected
requirements of those people who will be using the network.
The third step is to identify the resources and constraints of the organization. Organization
resources that can affect the implementation of a new LAN system fall into two main categories
- computer hardware and software resources, and human resources. You must document an
organization's existing computer hardware and software, and identify and define its projected
hardware and software needs. The answers to some of these questions will also help you
determine how much training will be required, and how many people will be needed to support
the LAN. The questions you ask should include:
Following these steps, and documenting the information in the framework of a formal report,
will help you estimate costs and develop a budget for the implementation of a LAN.
8.1.3 General network design process
In technical fields, such as engineering, the design process includes:
One of the methods you can use in the process of creating a design is the problem solving
cycle. This is a process that you use repeatedly until you finish a design problem.
One of the methods that engineers use to organize their ideas and plans when doing a design
is to use the problem-solving matrix. This matrix lists alternatives and various choices, or
options, from which you can choose.
We are only going to follow those boxed in red.
engineering journal
logical topology
physical topology
cut sheets
problem-solving matrices
labeled outlets
labeled cable runs
summary of outlets and cable runs
summary of devices, MAC addresses, and IP addresses
You might also ask your instructor if there is any other documentation that is relevant to your
project. Perhaps the most important part of the network design process is designing according
to the ANSI/EIA/TIA and ISO/IEC industry standards. For an excellent introduction to these
standards (with PDF downloads available), see the Siemon Company Guide to Industry
Standards @ http://www.siemon.com/standards/homepage.html.
8.2.1 Overview of wiring closet selection
One of the early decisions you must make when planning your network is where to place the
wiring closet(s), since this will be where you will have to install many of the networking cables
and networking devices. (Note: Detailed examples and practice for wiring closets are
provided.) The most important decision is the selection of the Main Distribution
Facility/Facilities (MDF). There are standards governing MDFs and IDFs, and you will learn
some of these standards while learning how to select the network wiring closet(s). If possible,
tour your own school's (or one of your local business') MDF/IDF.
Finally, you will learn how to plan your network so that you can avoid some of the problems
related to negative effects on networks caused by AC electricity from the power company.
8.2.2 Size
TIA/EIA-568-A specifies that, in an Ethernet LAN, the horizontal cabling runs must be attached
to a central point in a star topology. The central point is the wiring closet, and is where the
patch panel and the hub must be installed. The wiring closet must be large enough to
accommodate all of the equipment and wiring that will be placed in it, and include extra space
to accommodate any future growth. Naturally, the size of the closet will vary with the size of the
LAN, and the types of equipment required to operate it. A small LAN needs only a space the
size of a large filing cabinet, while a massive LAN requires a whole room.
TIA/EIA-569 specifies that each floor must have a minimum of one wiring closet and that
additional wiring closets should be provided for each 1,000 m2, when the area of the floor that
is served exceeds 1,000 m 2, or the horizontal cabling distance exceeds 90 m.
8.2.3 Environmental specification
Any location that you select for a wiring closet must satisfy certain environmental requirements
that include, but are not limited to, power supply and heating/ventilation/air conditioning
(HVAC) issues. In addition, the location must be secure from unauthorized access, and must
meet all applicable building and safety codes.
Any room, or closet that you choose to serve as a wiring closet should adhere to guidelines
governing such items as the following:
A minimum of two walls should be covered with 20mm A-C plywood that is at least 2.4m high.
If the wiring closet serves as the MDF for the building, then the telephone point of presence
(POP), may be located inside the room. In such a case, the interior walls of the POP site,
behind the PBX, should be covered from floor to ceiling with 20mm plywood, with minimum of
4.6 m of wall space provided for the terminations and related equipment. In addition, fire
prevention materials that meet all applicable codes (e.g. fire-rated plywood, fire-retardant paint
on all interior walls, etc.) should be used in the construction of the wiring closet. Rooms must
not have a dropped, or false, ceiling. Failure to observe this specification could result in an
insecure facility, allowing possible unauthorized access.
8.2.5 Temperature and humidity
The wiring closet should include sufficient HVAC to maintain a room temperature of
approximately 21° C, when all LAN equipment is in full operation. There should be no water or
steam pipes running through or above the room, with the exception of a sprinkler system,
which may be required by local fire codes. Relative humidity should be maintained at a level
between 30%-50%. Failure to adhere to these particular specifications could result in serious
corrosion of the copper wires that are contained within the UTP and STP. Such corrosion
would deter efficient functioning of the network.
8.2.6 Lighting fixtures and power outlets
If there is only one wiring closet in a building, or if the closet serves as the MDF, it should have
a minimum of two dedicated, non-switched, AC duplex electrical outlet receptacles, each on
separate circuits. It should also have at least one duplex power outlet positioned every 1.8 m
along each wall of the room, and should be positioned 150 mm above the floor. A wall switch,
that controls the room’s main lighting, should be placed immediately inside the door.
While fluorescent lighting should be avoided for cable pathways because of the outside
interference that it generates, it can be used in wiring closets with proper installation. Lighting
requirements for a telecommunications closet specify a minimum of 500 lx (brightness of light
equal to 50 foot candles), and that light fixtures be mounted a minimum of 2.6 m above the
floor.
8.2.7 Room and equipment access
The door of a wiring closet should be at least .9 m wide, and should swing open out of the
room, thus ensuring an easy exit for workers. The lock should be located on the outside of the
door, but allow anyone who is on the inside to exit at any time.
A wiring hub and patch panel may be mounted to a wall with a hinged wall bracket, or with a
distribution rack. If the choice is a hinged wall bracket, the bracket must be attached to the
plywood that covers the underlying wall surface. The purpose of the hinge is to allow the
assembly to swing out so that workers and repairmen can easily access the back side of the
wall. Care must be taken, however, to allow 48 cm for the panel to swing out from the wall.
If the choice is a distribution rack, then it must have a minimum 15.2 cm of wall clearance for
the equipment, plus another 30.5-45.5 cm for physical access by workmen and repairmen. A
55.9 cm floor plate, used to mount the distribution rack, will provide stability, and will determine
the minimum distance for its final position.
If the patch panel, hub and other equipment are mounted in a full equipment cabinet, they
require at least 76.2 cm of clearance in front, in order for the door to swing open. Typically,
such equipment cabinets are 1.8 m high x .74 m wide x .66 m deep.
8.2.8 Cable access and support
If a wiring closet serves as an MDF, all cable running from it - to IDFs, computers, and
communications rooms on other floors of the same building - should be protected by 10.2 cm
conduit or sleeved core. Likewise, all such cable running into the IDFs should be run through
the same 10.2 cm conduit or sleeved cores. The exact amount of conduit that is required is
determined by the amount of fiber optic, UTP, and STP cable that must be supported in each
wiring closet, computer, or communications room. Care should be taken to include additional
lengths of conduit in order to provide for future growth. To meet this specification, a minimum
of two excess sleeved cores or conduits should be kept in each wiring closet. Where
construction permits, all conduit and sleeved core should be kept to within 15.2 cm of the walls.
All horizontal cabling that runs from work areas to a wiring closet should be run under a raised
floor. When this is not possible, the cabling should be run through 10.2 cm sleeves that are
placed above door level. In order to ensure proper support, the cable should run from the
sleeve directly onto a 30.5 cm ladder rack in the room. When used in this manner, to support
cable, the ladder rack should be installed in a configuration which supports the equipment
layout.
Finally, any wall/ceiling openings that provide access for the conduit, or sleeved core, must be
sealed with smoke and flame-retardant materials that meet all applicable codes.
8.3.1 Topology as floor plan
TIA/EIA-568-A specifies that when using an Ethernet star topology, every device that is part of
the network must be connected to the hub by a run of horizontal cabling. The central point of
the star topology, where the hub is located, is called the wiring closet. It helps to think of the
hub as the center point of a circle which has lines of horizontal cabling radiating from it, like
spokes from the center of a wheel.
In order to determine the location of a wiring closet, begin by drawing a floor plan of the
building (approximately to scale), and adding to it all of the devices that will be connected to
the network. As you do this, remember that computers are not the only devices that you will
want to connect to the network; there are also printers and file servers to consider.
When you have completed this process, you should have a floor plan that is similar to the one
shown in the Figure .
8.3.2 Selecting potential locations
A good way to start looking for a potential wiring closet location is to identify secure locations
that are close to the POP. The selected location can serve as either the sole wiring closet, or
as the MDF, if IDFs are required. The POP is where telecommunications services, provided by
the telephone company, connect to the building's communication facilities. It is essential that
the hub be located near it, in order to facilitate wide area networking and connection to the
Internet.
In the floor plan graphic, five potential locations for wiring closets have been selected. They are
marked on the graphic as A, B, C, D, and E.
8.3.3 Determining number of wiring closets
After you have drawn in all of the devices that are to be connected to your network (floor plan),
the next step is to determine how many wiring closets you will need to serve the area covered
by the network. You will use your site map to do this.
Use your compass to draw circles that represent a radius of 50 m. from each of the potential
hub locations. Each of the network devices that you drew on your floor plan should fall within
one such circle. However, if each horizontal cabling run can only be 90 m. in length, can you
think of a reason why circles with a radius of only 50 m. would be used?
After you have drawn the circles, look at the floor plan again. Are there any potential hub
locations whose catchment areas substantially overlap? If so, you could probably eliminate one
of the hub locations. Are there any potential hub locations whose catchment areas can contain
all of the devices that are to be connected to the network? If so, then one of them could
probably serve as the wiring closet for the entire building. If you will need more than one hub in
order to provide adequate coverage for all of the devices that will be connected to the network,
check to see if one of them is closer to the POP than the other(s). If so, you will probably want
to select it to serve as the MDF.
8.3.4 Identification practice
Use the floor plan provided in this lesson. Notice that there are five potential locations for wiring
closets indicated on the floor plan - A, B, C, D, and E. Using the scale indicated on the floor
plan, set the compass so that it will mark a circle that equals 50 m in diameter. Mark circles for
each of the potential wiring closet sites. Then answer the following questions:
Potential locations for wiring closets have already been identified. They are marked on the floor
plan as A, B, C, D, E, F, G, H, I, and J.
There is no heating or cooling duct in the dropped ceiling space of this room, nor is there a
heating or cooling duct into the room. The nearest thermostat for this section of the building is
located in Room 120. The walls are concrete block construction covered with asbestos.
Although the room has a lock, it can only be unlocked from the outside. Currently, the room
serves as the mail room for the building.
8.4.5 Closet D
Also centrally located, room D is slightly larger than room C. Its dimensions are approximately
2.4 m wide x 3 m deep. In addition, room D is near the POP. The room does not have a
dropped ceiling. A heating and cooling duct that passes through the top of the room is also
vented into the room. Temperatures inside the room are controlled by a thermostat that is
located just inside the door. The exit door swings outward, and is 2.4 m wide.
The floor is covered with ceramic tiles. Lighting is provided by an incandescent lighting fixture
in the ceiling. The light switch that turns the light on and off is located just outside the door.
There are eight electrical outlets in the room, two along each wall. The walls are of concrete
block construction and are painted with a fire-retardant paint. Currently, the room is used to
store extra office equipment, and is kept locked. The door can only be unlocked from outside
the room.
8.4.6 Closet E
Also centrally located in the building, room E is adjacent to the POP. Room E is smaller than
room D. Its dimensions are approximately 2.4 m wide x 1.5 m deep. A water pipe enters the
building through room E, and travels from there to other locations throughout the building.
There is also a hot water heater in room E. In spite of repeated attempts to remedy the
problem, the water pipes in room E are heavily corroded. There is no false ceiling in the room.
The floor is covered with ceramic tile. A heating and cooling duct that passes through the top of
the room is also vented into the room. The nearest thermostat is located in the corridor outside
the room.
Lighting is provided by an overhead incandescent light suspended from the ceiling. The switch
that turns the light fixture on and off is located just inside the door to room E. The door, which
is approximately .9 m wide, swings into the room when it is opened. There are two electrical
outlets in the room. They are located on opposite walls. Because of its contents, room E is kept
locked, and can be unlocked from either inside or outside the room.
8.4.7 Closet F
Room F is centrally located near the front of the building, next to the main entry, and behind
the receptionist's desk. Currently it is used as a cloak room. There are two doors into the room.
Each door is approximately .9 m wide, and each swings out when opened. Neither door has a
lock. Lighting is provided by an incandescent light fixture. There are two light switches that turn
the overhead light on and off. They are located just inside each door.
There are no heating or cooling vents into the room. The nearest thermostat is located along
the corridor wall outside Room 118. The floor is carpeted. The room has one electrical outlet. It
is located along the wall behind the receptionist's desk in the lobby. Also, Room F has high
voltage power lines running through its outside walls.
8.4.8 Closet G
Room G is relatively small. Its dimensions are approximately 1.8 m wide x .9 m deep. The
outside wall for room G is only a partial wall. It does not reach all the way to the 3.7 m high
dropped ceiling. It only extends from the floor, and is of drywall construction. The two back
walls do extend all the way to the dropped ceiling and are of concrete block construction. One
electrical outlet is located along the longer of the two back walls. Room G does not possess its
own lighting fixture. Lighting is provided by fluorescent lighting fixtures in the corridor, and in a
shared work space. There is no door into room G, however, the entryway is .9 m wide.
The floor is carpeted. There are no air vents from the heating and cooling duct into room G.
The nearest vent is located approximately 4.6 m away. The nearest thermostat is located on
the wall opposite the entry into room G. Currently, the space provided by room G houses the
water cooler, a small microwave, and a small refrigerator.
8.4.9 Closet H
Potential wiring closet H is a little larger than room G. Its dimensions are approximately 2.4 m
wide x .9 m deep. Although its door is .9 m wide, entry into room H is through a small narrow
hallway. When the door opens, it swings into the room. Water pipes run through the dropped
ceiling space of the room. High voltage electrical conduits also pass through the room. Lighting
is provided by an overhead incandescent light; however, the switch that turns the light on and
off is located outside the doorway into the room. The floor is carpeted. There is no heating or
cooling vent into the room, nor does any heating and cooling ductwork pass through the
dropped ceiling space of this room. The nearest thermostat is located in the main corridor,
around the corner. There is just one electrical outlet in room H.
8.4.10 Closet I
Potential wiring closet I is located in the far corner of the building, next to the main entry. Its
dimensions are approximately 2.4 m wide x 4.6 m deep. Room I houses the heating and
cooling equipment for the building. All heating and cooling ducts to other parts of the building
lead from this room. High voltage electrical conduit passes through this room along the outside
walls. All walls are of concrete block construction, and are covered with fire-retardant paint.
The room does not have a dropped ceiling. The floor is covered with ceramic tile. Lighting is
provided by an overhead incandescent lighting fixture. The switch that turns the light on and off
is located just inside the door. When the door is opened, it swings outward. Because the room
houses potentially dangerous equipment, the door locks, and can be unlocked from either
inside or outside the room.
8.4.11 Closet J
Potential wiring closet J is located at one end of the building. Its dimensions are approximately
.9 m wide x 2.4 m deep. High voltage power lines enter the building through room J. High
voltage electrical conduit leads from room J to other critical areas of the building. The floor is
tiled. There is a dropped ceiling. The door is .9 m wide, and swings out when opened. Because
it is equipped with potentially dangerous equipment, the door to the room is kept locked. The
door can be unlocked from either inside or outside the room.
Lighting is provided by an overhead incandescent lighting fixture. The switch that turns the light
on and off is located inside the doorway, on the right side. There are two electrical outlets in
the room, and are located along opposite walls. All walls are of concrete block construction,
and are covered with fire-retardant paint. A heating and cooling duct passes through the
dropped ceiling space above the room, but there is no vent outlet into the room.
8.5.1 Catchment area problems
If the 100 m catchment area of a simple star topology wiring closet cannot provide enough
coverage for all the devices that need to be networked, the star topology can be extended by
using repeaters. Their purpose is to avoid the problem of signal attenuation, and are called
hubs. Generally speaking, when repeaters, or hubs, are used in this manner, they are located
in additional wiring closets called IDFs. These IDF closets are linked by networking media to a
central hub located in another wiring closet called the MDF. TIA/EIA-568-A specifies the use of
one of the following types of networking media:
The TIA/EIA recommends the use of CAT 5 UTP for horizontal cabling, when an Ethernet LAN
uses a simple star topology.
8.5.2 MDF location in a multi-story building
The main hub of an extended star topology Ethernet LAN is usually centrally located. This
central location is so important that in a high rise building, the MDF is usually located on one of
the middle floors of the building, even though the POP might be located on the first floor, or in
the basement.
The main graphic illustrates where the backbone cabling and horizontal cabling would be used
in an Ethernet LAN, in a multi-story building. In the Figure to the left, the backbone cabling (red
lines) connects the POP to the MDF. Backbone cabling is also used to connect the MDF to the
IDFs located on each floor. Horizontal cabling runs (blue lines) radiate out, from the IDFs on
each floor, to the various work areas. Wherever the MDF is the only wiring closet on the floor,
horizontal cabling radiate from it to the PCs on that floor.
8.5.3 Example of where you would use multiple wiring closets
Another example of a LAN that would probably require more than one wiring closet would be a
multi-building campus. The main figure illustrates locations where backbone and horizontal
cabling have been placed, in an Ethernet LAN, in just such a multi-building campus. It shows
an MDF in the center of the campus. In this instance, the POP is located inside the MDF. The
backbone cabling (red lines) runs from the MDF to each of the IDFs. The IDFs (yellow boxes)
are located in each of the campus buildings. In addition, the main building has an IDF, as well
as an MDF, so that all computers fall within the catchment area. Horizontal cabling, running
from the IDFs and MDFs to the work areas, is represented by the blue lines.
8.5.4 Cabling for MDF and IDF connections
The type of cabling that TIA/EIA-568 specifies for connecting wiring closets to each other, in an
Ethernet LAN extended star topology, is called backbone cabling. Sometimes - to differentiate
it from horizontal cabling - you may see backbone cabling referred to as vertical cabling.
In the extended star topology, there are two ways in which an IDF can be connected to the
MDF. In the first, each IDF can be connected directly to the main distribution facility. In this
case, because the IDF is where the horizontal cabling connects to a patch panel in the wiring
closet, whose backbone cabling then connects to the hub in the MDF, the IDF is sometimes
referred to as the horizontal cross-connect (HCC). The MDF is sometimes referred to as the
main cross-connect (MCC) because it connects the backbone cabling of the LAN to the
Internet.
A second method of connecting an IDF to the central hub uses a "first" IDF interconnected to a
"second" IDF. The "second" IDF is then connected to the MDF. The IDF that connects to the
work areas is called the horizontal cross-connect. The IDF which connects the horizontal
cross-connect to the MDF is called the intermediate cross-connect (ICC). Note that no work
areas or horizontal wiring connects to the intermediate cross-connect when this type of
hierarchical star topology is used.
When the second type of connection occurs, TIA/EIA-568-A specifies that no more than one
ICC can be passed through to reach the MCC.
8.5.7 Maximum distances for backbone cabling
As you have already learned, the maximum distances for cabling runs varies from one type of
cable to another. For backbone cabling, the maximum distance for cabling runs can also be
impacted by how the backbone cabling is to be used. To understand what this means, assume
that a decision has been made to use single-mode fiber-optic cable for the backbone cabling. If
the networking media is to be used to connect the HCC to the MCC, as described above, then
the maximum distance for the backbone cabling run would be 3,000 m.
At times the maximum distance of 3000m for the backbone cabling run must be split between
two sections. This occurs when the back bone cabling is to be used to connect the HCC to an
ICC and the ICC to the MCC. When this occurs, the maximum distance for the backbone
cabling run between the HCC and the ICC is 500 m. The maximum distance for the backbone
cabling run between the ICC and the MCC is 2,500 m.
The Figure lists TIA/EIA-568-A specifications for maximum distances for backbone cabling
runs for each type of networking media.