01.

INTRODUCTION TO COMPUTER NETWORK

All computers perform three basic tasks: input,

processing, and output. Some components of
02. NETWORK HARDWARE ESSENTIALS
computers are designed to perform only one of
these three functions; others are designed to
perform two or all three functions.
Network repeaters and hubs take incoming bit
Storage is a major part of a computer’s signals and repeat them at their original strength
configuration. Storage types include short-term out all connected ports. A hub is just a multiport
storage (RAM) and long-term storage (disk drives repeater. Hubs are a central connecting device for
and USB or flash drives). Data is stored as bits. A multiple computers, but because hubs allow only
collection of 8 bits is a byte. When expressing an one device to communicate at a time, the
amount of computer data, you can use bits or bandwidth of each port must be shared between
bytes. Bits are expressed as a lowercase b while all connected computers.
bytes are expressed as an uppercase B. Network interface cards create and mediate the
connection between the computer and the
PC hardware consists of four major components: a network medium. A computer’s MAC address is
motherboard, a hard drive, RAM, and firmware. defined on the NIC as a burned-in address. The NIC
The motherboard is the nerve center of the reads each frame arriving on the network medium
computer and contains the CPU, expansion slots, and determines whether the frame’s destination
and RAM slots. address matches its MAC address. If it matches or
is a broadcast frame, the NIC processes the frame;
The components needed to make a standalone otherwise, it’s discarded.
computer a networked computer include a NIC, a Switches use switching tables to determine which
network medium, and usually an interconnecting MAC address can be found on which port. Switches
device. In addition, network software consisting of can operate in full-duplex mode, allowing
client and server software, protocols, and the NIC connected devices to both transmit and receive
driver are needed to enable a computer to data simultaneously. Hubs operate only in
communicate on a network. half-duplex mode.
The layers of the network communication process Access points are central devices in wireless
can be summarized as user application, network networks and perform similar functions to hubs.
software, network protocol, and network interface. An AP requires devices to use an RTS signal when
they want to transmit data, and the AP responds
The terms for describing networks of different with a CTS signal when it’s okay to transmit. This
scopes are LAN, internetwork, WAN, and MAN. A extra network traffic reduces the effective
LAN is a single collection of devices operating in a bandwidth of wireless networks.
small geographic area. An internetwork is a Network switches interconnect multiple
collection of LANs tied together by routers, and a computers, just as hubs do. However, instead of
WAN and MAN are geographically dispersed simply regenerating incoming bit signals and
internetworks. repeating them to all other ports, a switch reads
Packets and frames are the units of data handled the destination MAC address in the frame,
by different network components. Packets, which determines which port the destination device is
are processed by the network protocol, are units of connected to, and forwards the frame to only that
data with the source and destination IP addresses port.
added. Frames, which are processed by the Wireless NICs perform the same function as wired
network interface, have MAC addresses and an NICs. Wireless NICs must be selected to match the
error-checking code added to the packet. wireless standard supported on the AP. When a
wireless client connects to an AP, it uses the SSID
to identify the wireless network’s name. Routers
 connect LANs to one another and forward packets The most common network technology for LANs is
from one LAN to another, according to the Ethernet. It’s described in IEEE 802.3 and has many
destination IP address specified in the packet. subcategories, including 10BaseT, 100BaseT, and
Routers use routing tables to determine where to 1000BaseT, that use twisted-pair copper cabling.
forward packets. Ethernet uses the CSMA/CD access method, which
Unlike hubs and switches, routers don’t forward is turned off when a full duplex connection is
broadcast frames. Each interface on a router is the established. Other Ethernet standards include
delimiter for a broadcast domain. When a router fiber-optic implementations, such as 100BaseFX
receives a unicast frame, it reads the destination IP and 1000BaseLX, among others. Faster standards
address and compares it with the list of networks such as 2.5GBaseT and 5GBaseT are compatible
in its routing table. If a match is found, the router with existing Cat 5e and Cat 6 cable plants,
forwards the packet to the destination network or whereas 10GBaseT requires Category 6a cabling.
to another router that gets the packet to its Wi-Fi is a wireless technology based on Ethernet,
destination. If no match is found, the router but it uses the CSMA/CA media access method.
discards the frame. If a router has a default route The most common Wi-Fi standards are 802.11b,
defined, it forwards any packets that don’t match 802.11g, 802.11a, 802.11n, and 802.11ac, with
networks in its routing table to the default route. speeds from 11 Mbps up to several Gbps. 802.11ax
will be the next 802.11 standard; it is still in
development.
03. NETWORK TOPOLOGIES AND TECHOLOGIES The antenna on a Wi-Fi device is both the
transmitter and receiver. Its characteristics and
Networks can be described by a physical and placement determine how well a device transmits
logical topology. The physical topology describes or receives Wi-Fi signals in an environment.
the arrangement of cabling that connects one Antennas are usually categorized by their radiation
device to another. The logical topology describes pattern: omnidirectional or unidirectional.
the path data travels between devices. The logical Wi-Fi signal interference can severely affect a
and physical topology can be different, and often WLAN’s performance and reliability. Common
are. types of interference include absorption,
The main physical topologies are the bus, star, ring, refraction, diffraction, reflection, and scattering.
and point-to-point. A physical bus topology is Noise from equipment and other wireless devices
simple but is no longer in common use because of and networks can also interfere with a Wi-Fi signal.
a number of weaknesses. A star topology, along This interference can cause signal degradation and
with the extended star, is the most common for errors that reduce the overall speed of data
implementing LANs. A physical ring topology isn’t transfers over a wireless network.
in widespread use now, but was used mainly in Token ring and FDDI are obsolete technologies that
network backbones. Point-to-point topologies are used a token-passing access method. Token ring
used primarily in WANs and with wireless bridges. operated at speeds of 4 Mbps and 16 Mbps and
Several point-to-point connections can create a ran over twisted-pair cabling, whereas FDDI ran
mesh topology for the purpose of redundancy. over fiber-optic cabling at 100 Mbps.
The main logical topologies are bus, ring, and
switched. A logical bus can be implemented as a
physical star or a physical bus and is used with 04. NETWORK MEDIA
hub-based Ethernet and Wi-Fi networks. A logical
ring can be implemented as a physical ring or a
physical star and is most commonly seen in token Network media are the materials through which
ring and FDDI networks. The switched topology network signals travel between devices. They can
uses a physical star and is used with Ethernet be a physical material, such as copper wire or glass
networks and a switch in the center of a star fiber, or simply the air. When a physical material is
physical topology. used as the medium, it’s usually referred to as
A network technology defines the structure of “wired networking,” and when signals are
frames and how a network interface accesses a transmitted through the air, the medium is aptly
medium to send frames. It often defines the media called “wireless networking.
types that must be used to operate correctly. Wired networking media come in two main
categories: copper and fiber optic. Cable
 characteristics include bandwidth rating, maximum TCP/IP is the main protocol suite used in networks.
segment length, susceptibility to interference and Like most facets of networking, TCP/IP takes a
eavesdropping, and cable grade. layered approach and is organized in these four
Twisted-pair cabling comes in shielded or layers: Application, Transport, Internetwork, and
unshielded varieties. Most networks use UTP, but Network access.
STP can be used in electrically noisy environments. The Application layer consists of protocols such as
Cat 5e and Cat 6 are the most common cable types HTTP and DNS and provides an interface for
in networks today. applications to access network services. Other
Twisted-pair cabling components consist of Application-layer protocols include POP3, IMAP,
connectors, patch cable, jacks, patch panels, and and SMTP for e-mail; FTP, TFTP, and SMB for file
distribution racks. A structured cabling plant transfer and sharing; Telnet and SSH for
consists of work areas, horizontal wiring, commandline remote access; and SNMP for device
telecommunications closets (IDFs), equipment management.
rooms (MDFs), backbone cabling, and entrance The Transport layer provides reliability and works
facilities. with segments (TCP) and datagrams (UDP).
Fiber-optic cable uses pulses of light to represent Transport-layer protocols are used with most
bits and is immune to EMI, RFI, and electronic Application-layer protocols because they supply a
eavesdropping. Commercial implementations of up header field to identify the Application layer and
to 100 Gbps are in use. Each network connection provide reliability and flow control for applications
requires two strands of fiber-optic cable: one for that typically transfer a large amount of data.
transmitting and one for receiving. Fiber-optic The Internetwork layer is where most network
cable comes in single-mode or multimode; configuration occurs and is composed of IP, ICMP,
single-mode uses lasers and can carry data longer and ARP. This layer is responsible for four main
distances, and multimode uses LEDs. tasks: defining and verifying IP addresses, routing
Wireless networks can be subdivided into LANs, packets through an internetwork, resolving MAC
extended LANs, and mobile computing. The addresses from IP addresses, and delivering
components of a wireless LAN are a NIC, an packets efficiently.
antenna, and a transceiver or an access point. The Network access layer is composed of network
Wireless networks send signals in the form of technologies, such as Ethernet and WAN
electromagnetic waves. Different network types technologies. Some tasks it performs include
use different frequencies for signal transmission. providing a physical (MAC) address for the network
Different technologies are used to transmit and interface, verifying that incoming frames have the
receive data, including infrared, laser, narrowband correct destination MAC address, receiving packets
radio, and spread-spectrum radio. Infrared can from the Internetwork layer and encapsulating
deliver speeds up to 100 Mbps and is used in some them to create frames, and transmitting and
LAN applications. Laserbased technologies require receiving bit signal.
line of sight between sender and receiver, as does
infrared, but laser isn’t as susceptible to 06. IP ADDRESSING
interference from other light sources. Narrowband
radio uses low-powered, two-way radio An IPv4 address is a 32-bit dotted decimal number
communication and is highly susceptible to separated into four octets. Every IP address must
interference. Spread-spectrum LANs are the most have a subnet mask to indicate which part of the IP
common and are used for 802.11 b/g/n Wi-Fi address is the network ID and which part is the
networks. host ID.
Criteria for choosing LAN media include needed There are three main address classes: A, B, and C.
bandwidth, budget, environmental factors, the Address classes determine the default network ID
distance the network must span, and the existing and host ID portions of an IP address. Each class
cable plant, if any. Networks combining fiber-optic, has a range of private IP addresses that can’t be
UTP, and wireless have become the norm. used on the Internet; they’re used to address
private networks.
05. NETWORK PROTOCOLS CIDR largely replaces the IP address class system; it
uses a prefix number or subnet mask to determine
the network and host IDs of an IP address.
 Subnetting enables an administrator to divide a The OSI reference model and IEEE Project 802
large network into smaller networks that require a define a frame of reference for networking and
router for communication. It also allows an ISP to specify the operation of most networking
allocate only the number of public IP addresses a technologies in current use. Together, these
company requires instead of assigning an entire models describe the complex processes and
address class, thereby conserving public IP operations involved in sending and receiving
addresses. Subnetting also allows dividing information across a network.
networks into logical groups for security and The OSI reference model separates networking
efficiency. into seven layers, each with its own purposes and
There are several rules for IP address assignment. activities. From the top down, the layers are
Only Class A, B, and C addresses can be assigned to Application, Presentation, Session, Transport,
a host; every IP address must have a subnet mask; Network, Data Link, and Physical. Most network
all host IDs on the same network must be unique; products and technologies are also specified in
you can’t assign the network number or broadcast terms of the layers at which they operate. The
address to a host; computers with different layers help describe the features and functions the
network IDs can only communicate through a products and technologies deliver.
router; and the default gateway must have the Following is a summary of the functions of each
same network ID as the host. OSI model layer:
Commands for working with IP address ❍ Application—Provides access to network
configurations include netsh, ipconfig, ping, arp, resources
route, tracert, and nslookup. ❍ Presentation—Handles data formatting and
Network Address Translation (NAT) enables an translation
organization to use private IP addresses while ❍ Session—Manages ongoing conversations
connected to the Internet. Port Address between two computers
Translation (PAT) allows several hundred ❍ Transport—Breaks long data streams into
workstations to access the Internet with a single smaller chunks (segments)
public Internet address. ❍ Network—Provides best path selection and IP
IPv6 will eventually replace IPv4 because of addressing
advantages such as a larger address space, a ❍ Data Link—Defines how computers access the
hierarchical address space, autoconfiguration, media
built-in QoS, and built-in security. IPv6 addresses ❍ Physical—Converts bits into signals and defines
are expressed as eight four-digit hexadecimal media and connectors
values. The IEEE 802 project defines networking standards
IPv6 defines unicast, multicast, and anycast in more than 20 categories to ensure that network
addresses. Unicast address types include link-local interfaces and cabling from different
addresses, unique local addresses, and global manufacturers are compatible. The IEEE 802.2
addresses. standard specifies the functions of a network’s
IPv6 can configure address settings automatically. Physical and Data Link layers by dividing the Data
IPv6 autoconfiguration occurs by using stateful or Link layer into two sublayers: Logical Link Control
stateless autoconfiguration. With stateless (LLC) and Media Access Control (MAC). Together,
autoconfiguration, the node listens for router these sublayers handle media access, addressing,
advertisement messages from a local router. With and access control and provide reliable, error-free
stateful autoconfiguration, the node uses an delivery of frames from one computer to another
autoconfiguration protocol, such as DHCPv6.
Transitioning an entire network from IPv4 to IPv6
successfully while maintaining compatibility with
IPv4 requires a variety of transition technologies,
including dual IP layer architecture, IPv6-over-IPv4
tunneling, ISATAP, 6to4, and Teredo.

07. NETWORK REFERENCE MODELS AND STANDARDS