Welcome to the OUTWARD course "Computer Network

Management"!
The estimated runtime of this course is 50 minutes.

1
Here you see how to navigate within the course.

2
3
The purpose of this course is to provide you with an
understanding of what may impact network processing
speed, and what TCP/IP is.
In addition, you will learn why it is important to know about it
and to understand what a network operating system is and
the tools available to diagnose network problems.

4
Think about your Internet service provider. If you were to
source a new contract for Internet service, what factors
would impact your decision?
When you use the Internet, or your local area network, are
there things that make it less convenient than you prefer?
How about things like network speed and network stability, or
perhaps reliability or security?
These are all aspects of network management, and this
course explains how to manage the basics of networks such
as speed, stability and problem solving.
In addition, we will introduce the operating system and
diagnostic tools to maintain an efficient and comfortable
network.

5
This lesson explains network speed and the circumstances
that affect network speed.

6
Network speed is the rate of data transmission between
network nodes.
The faster the speed, the more data can be transmitted in a
given period of time.
For units of speed, bps is used, which indicates the number
of bits of data transferred in a second, and Kbps or Mbps are
frequently used nowadays.
Speed is largely affected by physical mediums such as
cables or wireless LAN technology.
Moreover, differences due to transmission types such as
digital transmission vs. analog transmission, or the condition
of the network also affects speed.

7
In a baseband network, such as Ethernet, a device transmits
digital signals directly without manipulation.
In the baseband system, the transmission method is
visualized using line cords.
Only one communication channel can be used, and therefore
only one device can transmit signals at a time.
It can be compared to a road with a single traffic lane.
In a broadband network, the digital signals are modulated to
the carrier wave. The modulated carrier wave is transmitted
to the receiver, which demodulates the signal to digital data
again.
The physical bandwidth of the cable is practically separated
into multiple channels and each channel has its own carrier
frequency.
In other words, a single cable can be divided by the carrier
frequency, so you can send data via multiple channels at the
same time, meaning you can transmit them at high speed. It
can be compared to a road with multiple traffic lanes.

8
In the context of the network, throughput means the
communication speed.
Communication speed is a necessary condition to maximize
network efficiency.
In order to maintain the maximum communication speed and
maximize the network efficiency, it is important that
throughput corresponds to the bandwidth. The throughput
can be compared to a car's speed.
When cars run at the maximum speed allowed for each road,
cars can use the roadway the most efficiently.
The other major factor in the speed of network response is
the amount of data that passes through the network at any
one time, referred to as the network traffic.
In large networks where many users are generating
messages simultaneously, more bandwidth is required to
maintain response speed.
To prevent system congestion, network operating systems
and protocols are designed to manage and distribute

9
network traffic.
This feature also prevents the transactions of one user from
degrading the performance of the network for other users.
Network operating systems and protocols can be compared
to the role of directing traffic.

9
Test your knowledge in a quiz!

10
Network speed is expressed by bps, which indicates the
number of bits of data transferred in a second.
The faster the speed, the more data can be transmitted in a
given period of time.
Network speed is not constant. The speed changes based
on various factors, such as the physical medium,
transmission method, and amount of traffic.

11
This lesson explains about using TCP/IP to connect network
nodes.
It also explains about the use of ports and utilities that help
with the management of networks.

12
TCP and IP are network protocols. TCP/IP is a general term
for protocols such as TCP and IP.
TCP/IP is the universal standard for network communication.
As noted on the screen, it is used to support the Internet by
facilitating communication between computers all over the
world and providing access to data and applications at
geographically distributed sites.
It also supports intranets, which are private networks only
available to authorized users, and extranets, which allow for
the sharing of data among multiple intranets.
TCP/IP is also used in virtual private networks, which are
networks where remote users can be given access rights to
establish a virtual point-to-point encrypted connection. The
virtual connection is connected via a dedicated tunnel over a
modem to the resources of a private network.

13
IP addresses are unique numbers assigned to each
computer or node on the network.
It is compared to a "network address" that is assigned to
recognize each computer.
There are two ways to describe network addresses, decimal
numbers and binary numbers.
Although the actual IP address is described as a binary
number, it can be described as a decimal number.
A 32-bit long number that is described as a binary number is
divided into four 8-bit groups and each group is converted to
decimal numbers.
This 8-bit unit is called an "octet".
There are two types of IP addresses, IPv4 and IPv6. IPv6
was created to support the extremely high number of IP
addresses expected to be needed in the near future. This
course deals with IPv4 addresses.
IP addresses that are valid in a limited area, such as inside a
company, are called "private IP addresses". Private IP

14
addresses do not allow nodes on the network to
communicate directly with the Internet. In addition, private IP
addresses may be used in other networks.

14
IP addresses are divided into a network ID and a host ID.
The division point is regulated by the class assignment of the
default net mask.
Class A is when the most significant bit is 0, and the leading
octet 0 to 127 applies.
In Class A, the leading 8 bits is defined as the network ID,
and it has 128 combinations of network addresses and
approximately 160,000 host addresses for each. They are
used for very large networks because the number of nodes
is enormous.
Class B is when the most significant bit starts from 10, and
the leading 16 bits is defined as the network ID.
They are used for large networks for the same reason.
Class C is when the most significant bit starts from 110, and
the leading 24 bits is defined as the network ID.
They are used for most organization networks.
Class D is for multicast group addresses used to send sound
or image data at once.

15
Class E is reserved for experimentation and development,
and is not in use.

15
Specific IP addresses are assigned for special purposes and
therefore cannot be used for other purposes.
An address starting with 0 references the local node within
the current network. For example, 0.0.0.17 references
device 17 in the current network.
127.0.0.1 references the local loopback inside a workstation,
and is used to test whether TCP/IP is successfully installed.
An address with 255 is the broadcast address.
255.255.255.255 sends a message to every node on the
Internet.
Moreover, when the address is 192.168.2.255/24, messages
are sent to the entire network belonging to 192.168.2.xxx.
When divided into subnets, the highest IP address of that
network is used.

16
As mentioned previously, IP addresses are separated into
the network ID and the host ID, but there are some
exceptions.
Therefore, subnetworks are used which separate the
Network ID and Host ID using a subnet mask.
In subnet networks there are cases when routers or
gateways physically separate segments or cases when they
are physically part of the network but separated logically.
An example is a Class C network; by expanding the subnet
mask up to 26 bits, 00, 01, 10 and 11 subnetworks can be
added in the expanded area.
As a result, 254 host users in 1 network are divided into 62
host users on 4 networks.

17
DHCP is used to automatically allocate IP addresses.
Clients acquire an IP address with DHCP by temporarily
setting their IP address to 0.0.0.0, and broadcasting a
message.
DHCP servers receive the message and send a message
with an available IP address.
Clients then receive the message and broadcast the
available IP address.
When DHCP servers response to the message, clients can
officially set up the IP address.
As DHCP can be used to automatically set up IP addresses,
it is effective when there are many terminals under
management.
However, one disadvantage of DHCP is that there may be
communication errors if different IP addresses are allocated
frequently or the DHCP server fails.
As DHCP is upward compatible with BOOTP, BOOTP is no
longer used.

18
Moreover, when DHCP cannot acquire an IP address, APIPA
may automatically allocate IP addresses.
However, as link local addresses will not be routed,
communication with other networks and the Internet is not
possible.

18
As memorizing IP addresses is difficult, it is possible to add
domains which have supported names.
Domains are acquired by combining a free name and a top-
level domain tailored to the intended use.
Furthermore, domains can be accessed using browsers and
the like by specifying protocols, servers and files to be
accessed, forming a URL.
Domains are managed by the DNS system. DNS servers
maintain domains and corresponding IP addresses.
When accessing a web server using a URL, first the DNS
server is asked for the domain name and the IP address is
acquired.
This procedure is called name resolution. You can access
necessary data by accessing the acquired IP address.
If the name is not resolved, the DNS server contacts the
name server. A name server exists for each domain
hierarchy.
The DNS server sends repeated inquiries from the top of the

19
hierarchy until target IP address is obtained.

19
TCP or UDP identifies the application transmitting data
across the network by a 16-bit port number attached to the
IP address.
UDP, an abbreviation for User Datagram Protocol, is used
when immediacy is more important than reliability regarding
data communication.
The IP address combined with the port number is called a
socket or socket address and written as ":25".
Put simply, port numbers indicate which process or service
to send the transmitting data to.
If it is compared to our world, the IP address is the location
of the apartment building and the port number is the
apartment number.
Port numbers 0 to 1023 are called well-known port numbers
and are mainly assigned to server applications.
For example, TCP/IP systems that function as an FTP server
all use port 21.
Port numbers 1024 or higher are called high ports and are

20
assigned to custom applications.
By combining the IP address with port numbers, one network
computer can access multiple different services at the same
time.
The client IP address and port number plus the server IP
address and port number are referred to as socket pairs and
they identify each TCP or UDP connection on the network.

20
The SNMP agent monitors the performance of the network
device and notifies the SNMP manager.
The SNMP protocol is used for communicating this
information.
TCP/IP provides for management of all links and nodes in
the network through the use of the SNMP agent in all
network components that feed back information to SNMP
managers.
For example, it is possible to require a response from an
address, or to notify transmission errors when addresses are
down.
The system is based on a management information
database, called MIB, that sets out the standards for
continued operation of each network component.
ICMP is another IP protocol used in network-layer
management and control, and to report network errors.
If a message cannot be delivered by a router, the router will
return it to the source with an ICMP message.

21
There are differences in the packet size that can be
transmitted, depending on the type of data link.
For example, Ethernet can transmit up to 1500 bytes, while
FDDI can transmit up to 4352 bytes.
FDDI stands for Fiber Distributed Data Interface and is a
network for large scale LANs to send data via optical fiber.
The size of the packet transmitted is called the MTU.
Furthermore, dividing the size of the packet into a size that
can be sent at one time is called IP fragmentation.
In order to understand the MTU value, Path MTU Discovery
is used.
If the packet size exceeds the upper limit, the router sends
an ICMP message back to the source.
When the workstation receives an ICMP message, it divides
and sends the packet data in a smaller size.
The work station measures the MTU by repeating this data
transfer. This process is called Path MTU Discovery.

22
Network utilities can help to investigate network performance
or resolve connection issues.
In Windows, the command prompt can be used for this
purpose.
Let us learn more about command prompts such as ping,
tracert, netstat, ipconfig and nslookup.

23
"ping" stands for Packet Internet Groper.
A ping command sends a diagnostic packet to a designated
network node to check the network connection.
If the node receives the packet, it responds, confirming that
the link is operational. If the node does not respond, the user
is alerted to a link failure.
Ping uses ICMP to send the request and return the response
or the fact that the message could not be delivered.
A network administrator will ping nodes to try to identify and
isolate problems on the network and to measure
performance.

24
The "tracert" command is used to perform a trace route and
it is a very useful diagnostic tool that goes hand in hand with
ping in diagnosing common network connection issues.
The tracert command is able to show the individual network
nodes a packet goes through to reach its destination.
If at any point connectivity is lost using tracert it is possible to
see up to what node communication was successful.
It might be that not all network nodes will return the ping.
The first number corresponds to the position of the node in
the network path from the client to the server.
The next three numbers indicate the time taken for three
echo requests to reach and bounce back from that
host/router.
Finally, the Fully Qualified Domain Name, or FQDN for short,
and Internet Protocol, or "IP", address, of the host/router is
shown.
ICMP or UDP is used in tracert.
Starting with Microsoft Windows 2000, a new tool called

25
"pathping" is also available and although it is similar to
"tracert" it computes better average response times for each
node on the network path.

25
The command "netstat" stands for network statistics.
It is useful mostly on servers when you want to ensure that a
particular process is running and it is correctly listening on
the network for incoming connections. The "netstat"
command has a number of parameters, the most commonly
used one being "netstat-an".
When running "netstat –an" you can see the protocol, port
number and IP address that each process is listening with on
the network.
Addresses shown as 0.0.0.0 indicate that this process is
listening on all available IP addresses of the host machine.

26
The "ipconfig" command stands for IP configuration and it is
used to display IP-level details of the machine.
By default, ipconfig returns a concise view that only shows
the DNS suffix, current IP address, subnet mask and default
gateway.
When troubleshooting, it is best to run ipconfig using
"ipconfig /all" in order to get a more complete listing of IP-
level details.
In this instance, in addition to the above, you will also be
able to see DNS, WINS and DHCP server details, as well as
the NICs MAC address.
In older versions of Windows, before the introduction of
"ipconfig", the command "winipcfg" returned similar results in
regard to the networking setting of the system.
The ifconfig command is used on Linux systems.

27
"nslookup" stands for Name Server LOOKUP.
It is possible to validate network problems, by contacting the
DNS server and diagnosing the IP address from the domain
name, and vice-versa.
For example, when a ping shows that connectivity exists but
the FQDN fails to resolve correctly to the IP address,
nslookup can be used to query and resolve the DNS server
used by the client machine.
In this manner, it is also useful in diagnosing problems with
the local DNS server.

28
Test your knowledge in a quiz!

29
There are classes and special purpose addresses in IP
addresses.
Some are not available, therefore use the appropriate
number depending on your purpose.
To make IP address management easier, there is an
automatic allocation method. Moreover, since IP addresses
are a series of numbers and are hard to remember, there is
a system called DNS which converts them into alphanumeric
characters.
Networks can effectively use limited IP addresses when they
are simply sorted into classes. For that reason, they may be
divided into subnetworks.
TCP/IP has a protocol to manage networks. They are widely
used with network utilities such as ping.

30
This lesson explains how network operating systems enable
workstations to use networks.

31
A network operating system, or NOS, is a computer
operating system that enables workstations to use network
resources and services.
In addition, a NOS makes such resources and services
available to the rest of the clients on the network via servers.
Examples of commonly used NOS are Mac OS, Windows
and Linux.
NOS respond to requests from users and applications, and
delivers a variety of services.
NOS enable access to files and resources, and provide file
sharing services. It can enable workstations and peripherals
to communicate with each other.

32
Recent developments in central processing units, commonly
known as CPUs, have allowed 64-bit processors to become
affordable and easily accessible for purchase by the general
public. 64-bit processors can offer significant advantages in
speed, the size of memory they are able to address and in
other areas and potentially deliver better performance.
However, operating systems that support 64-bit processors
are not automatically more secure or stable.
Moreover, unless the operating system and applications that
run on these processors are 64-bit enabled they are not able
to gain any advantages.
Microsoft Windows operating editions have provided 64-bit
support since Windows XP and Windows Server 2003.
Windows 64-bit systems support 32-bit applications as well
for backwards compatibility. 32-bit drivers are not supported,
therefore all drivers used in Windows 64-bit systems must
also be 64-bit.
Windows Server 2012 is Microsoft’s operating system for

33
servers.
It is an OS for servers based on Windows 8, and is a
successor OS to Windows Server 2008 R2.
Windows Server 2012 is based on Windows 8.1.
As server managers have been redesigned, managing a
number of servers is more convenient, and unless Server
Core mode is installed, modern UI are available.
Unlike Windows Server 2008 R2, Windows Server 2012 can
switch over without reinstalling Server Core and GUI install.

33
UNIX is a multi-user, multi-tasking operating system that is
widely used as the master control program in workstations
and servers. There are many versions of UNIX on the
market. Many IBM mainframes also run UNIX applications.
Because UNIX interfaces have been added to the operating
system, they have obtained UNIX branding such as MVS
and OS/390.
Although the majority of general computers are Windows,
UNIX is widely used in mission critical applications for
client/server and transaction processing systems.
Gnu/Linux is also built on the same principles as UNIX and
utilizes a similar architecture.
The UNIX vocabulary is exhaustive, with more than 600
commands that manipulate data and text.
Linux is one of the operating systems that is compatible with
UNIX.
It is written without diverting other OS and published as
open-source software.

34
Linux has a similar specification to UNIX, operates lightly, is
widely used and continues to grow in popularity.

34
UNIX has a sophisticated set of printing utilities that are
sometimes used under other operating systems.
They are called the LPR system and are TCP/IP based
protocol.
In this system, data to be printed is "spooled" to a file, where
it waits until the printer is free to print. This is known as print
queuing.
The queue is scanned by the line printer daemon, which acts
as the print server.
The benefit of the LPR system is that it quickly moves the file
to the spool area, freeing up the operating system to carry
out other tasks, while the LPD manages the printing.
LPR is used with TCP/IP under NOS such as Windows and
Mac OS to increase print speed.
However, today most Linux and some UNIX systems use the
Common UNIX Printing System, CUPS for short, instead of
LRP/LPD.
CUPS is an open-source printing system modularized for the

35
UNIX OS. Users can use CUPS with a web browser.

35
Apple's Mac OS X is designed to run only on Apple’s line of
hardware. High stability is ensured by creating hardware and
OS in one company.
A window system that is established in an original user
interface called Aqua is built-in and offers intuitive operation.
Mac OS was developed based on UNIX, so this is also a
robust OS like UNIX and has comprehensive command-line
tools.
Apple now supports TCP/IP as the base communication
protocol for Mac OS X, unlike some of its earlier operating
systems.
As a result Mac OS X machines can co-exist on the same
network as Linux and Windows workstations, and connect to
the same file and print servers and to one another easily.

36
OS X Server is a server operating system only for Macintosh,
developed and sold by Apple.
Basic parts such as the operating screen are the same as
Mac OS X, and include open source and server software.
In addition to basic server functions such as Web server,
mail server, and directory service, original functions such as
Wiki Server and iCal Server, Podcast Producer, Spotlight
Server are available as well.

37
Directory services hold various information such as user
names and passwords to use the network, machine names,
and the like, and LDAP is a popular communication protocol.
Typical directory services include NetIQ eDirectory, Apple
Open Directory, Microsoft Active Directory.
LDAP is a communication protocol to access directory
service using TCP/IP networks.
Microsoft Active Directory is a directory service built in to
Windows server since Windows 2000, and is used to
centrally manage various data or authority on resources and
users on the network.
In addition to the directory service, it also provides support
for user authentication and client management.
NetIQ eDirectory is a hierarchical object-oriented database
that represents every asset in the organization in a logic tree.
Assets include people in the enterprise, network devices,
network applications, relationships between information, and
so on.

38
By using dynamic authority inheritance and equivalence,
detailed control can be implemented efficiently.

38
Test your knowledge in a quiz!

39
Currently, operating systems include both 32-bit and 64-bit
versions.
64-bit processors provide the maximum amount of memory
usage, but users need to be cautious in terms of
compatibility aspects.
Windows, Macintosh, UNIX are commonly used network
operating systems.
Each product provides different features, therefore each
plays an active roll in different scenes.
Besides UNIX, which plays an important role in large
companies and financial systems, an advanced printing
utility called LPR are adopted for use with other NOS.
Moreover, by adopting directory services such as NetIQ
Directory or Microsoft Active Directory, it is possible to
manage network user names, passwords, and machine
names in a centralized way.

40
Ongoing network maintenance is essential for efficient
network performance.
Networks require both hardware and software maintenance,
which in turn requires a range of different network tools.
This section explains typical network tools, including cable
testers, protocol analyzers, network testers and Wireshark.

41
A cable tester is essential for analyzing considerable
amounts of cabling and numbers of connections.
They measure voltage and resistance levels at as many
nodes as possible around the network.
Voltmeters are instruments that measure system voltages
and inform the engineer if network signals are at appropriate
levels.
Ohmmeters measure the resistance of a device in ohms.
Ohmmeters typically test for open and short circuits in
network hardware, thus informing the engineer of the
integrity of connections between cabling and network
hardware.
Multimeters are devices that can be used to test voltage or
resistance and can be used to test for open circuits or
signals within networks.
By measuring the resistance between two connectors, it is
possible to test for open circuits or bad cable connections.

42
Protocol analyzers are software testers for tracking and
interpreting network protocols.
Protocols are essential for the efficient operation of networks
and provide information on interactions between
workstations, servers and peripheral devices. Therefore,
protocol analyzers can help diagnose intricate problems of
incompatibility between different network devices.
Because of stable standards that define LAN and WAN
interconnectivity and communication, there are very few
protocol incompatibilities in software and hardware
components that are shipped for use by businesses.
This has lead to the diminishing use of protocol analyzers by
everyday network technicians and administrators.
However, they remain extremely important and useful tools
for network designers and other IT specialists.

43
A network tester decodes and analyses data transmitted
over a network.
There are numerous types of testers available, ranging from
relatively simple software programs to purpose specific
workstations.
Testers not only receive existing network data, they can also
formulate and transmit data packets designed to test specific
aspects of the network.
Most network testers are software-based, typically running
on a laptop PC.
However, there are still some that are hardware-based, such
as the one shown here.

44
Wireshark is network analyzer software used to analyze and
display data flowing across the network.
Computers running Wireshark can collect and record
network packets, analyze protocols, display unique control
information for each protocol, and can be used to investigate
the cause of network faults and trouble.
Wireshark has a function to analyze more than 600 protocols,
including IP and DHCP, and supports many other file formats
such as Sniffer Pro and tcpdump.

45
Test your knowledge in a quiz!

46
To solve network problems, diagnostic tools are necessary to
understand the cause.
For example, a cable tester can be used if there could be a
problem with cabling, or a protocol analyzer if there could be
a problem with a protocol.
It is necessary to identify the problem according to the
situation and use appropriate tools to specify the cause.

47
This is the end of the course. Let us look back what you
have learned in this course.
Transmission speed is a very important factor when using a
network.
Speed is affected by various factors, therefore, you must pay
attention to show optimal performance when managing
networks.
TCP/IP, which controls communication, is extremely
important in network management.
It is necessary to grasp the usage of IP addresses and
management in TCP correctly, and to diagnose or solve
problems by using appropriate commands for the situation.
Workstations can be connected to a network by introducing a
network OS.
It is necessary to grasp the characteristics of various
operating systems and manage them appropriately.
Diagnostic tools may be helpful in understanding network
issues.

48
A variety of diagnostic tools are available, ranging from tools
to understand electrical problems with cables, to items that
can be used to analyze information in units of packets.

48
Congratulations! You have now completed the OUTWARD
course on computer network management.

49