Chapter 2

---

Basic Architecture
of Computer Networking

In the globalized world in which we live, the use of technologies is essential, as they make our daily tasks easier. In this environment, where we need to interact with each other constantly, we rely on a variety of communication resources that interconnect various electronic devices and give us quick and accurate answers, meeting our desires.

This chapter is divided into six sections that present the main networking architecture knowledge required for CCNA.

Data Communication

As Forouzan (2006), data communication is the exchange of information between two devices through a communication medium such as a wire pair.

A basic data communication system consists of five elements:

1. Message: the information to be transmitted. It may consist of text, numbers, pictures, audio, and video - or any combination of these elements;
2. Transmitter: is the device that sends the data message. It can be a computer, a workstation, a phone, a video camera, among others;
3. Receiver: it is the device that receives the message. It can be a computer, a workstation, a phone, a video camera, etc .;
4. Medium: is the physical path through which travels a message addressed to the receiver;
5. Protocol: is a set of rules governing the communication of da-. It is an agreement between devices that communicate.

Streaming Data

According to Torres (2004), there are three types of data transmission:

1. Simplex: In this type of data transmission, one device is the transmitter, and the other is the receiver. Simplex data transmission is therefore unidirectional;
2. Half-duplex: This type of data transmission is bidirectional, but because they share the same communication channel, the devices do not transmit and receive data at the same time;
3. Full-duplex: it is true two-way communication. A and B may transmit and receive data at the same time.

History

As Morimoto (2008c), the networks have gone through a long process of evolution before they reach the standards currently used. The first computer networks were also created during the 60s, as a way to transfer information from one computer to another.

A brief timeline shows some important moments of developing computer networks, as can be seen below.

60 - The Beginning

From 1969 to 1972, it was created ARPANET, the embryo of the Internet we know today. The network went live in December 1969, initially with only four of us, who responded by SRI names, UCLA, UCSB and Utah and were hosted, respectively, at the Stanford Research Institute, the University of California, the University of Santa Barbara and the University of Utah, all of them in the US. They were linked by 50 kbps links created using dedicated phone lines, adapted for use as a data link (MORIMOTO, 2008b, [unpaged]).

The main ARPANET network characteristics were:

a)     terminals "dumb" (without processor);

b)     communication with a central computer;

c)      consolidation of data communication principles;

d)     modem appearance;

e)    perception by the industry that the use of remote computers would be decisive in the following decades;

f)       individual investment of each manufacturer to develop its own teleprocessing technology;

g)     the huge growth of teleprocessing networks;

h)     geographic expansion;

i)        variety of applications;

j)       the emergence of the need for users of an access system applied from other systems;

k)     interconnection teleprocessing systems;

l)        computer networking.

Project ARPA

In 1974, TCP / IP emerged, which became the vo definitive protocol for use on ARPANET and later the Internet. A network linking several universities allowed free traffic information, leading to the development of resources that we USA- today, such as e-mail, telnet, and FTP, that allowed connected users to exchange information, access other computers remotely, and share files. At the time, mainframes with good processing power were rare and incredibly expensive, so they ended up being com- shared between several researchers and technicians who could be located anywhere on the network (MORIMOTO, 2008b, [unpaged]).

The main features of this network were:

a)    the early era of computer network technology;

b)     distributing applications across multiple interconnected computers;

c)     the teleprocessing systems continued to exist; however, each network computer had its own teleprocessing structure;

d)     packet switching;

e)    the division into several functional layers of the communication tasks between different computer applications;

f)       creating the basic concept of Computer Network Architecture;

g)     creating transport protocols;

h)    elaboration of mechanisms for flow control, reliability, and routing;

i)        development and operation of the first application protocols:

• FTP - File Transfer Protocol;
• TELNET - Virtual Terminal;

j)        interconnection American universities computers;

k)     interconnection of computers in other countries;

l)       opening a new market for companies specializing in the sale of telecommunications services: the provision of data communication services through the provision of a communication structure;

m) standardization of public packet networks from the development in 1976, the first version of Recommendation X.25

Network Concept

According to Sousa (1999), "computer network is a set of interconnected devices to exchange information and share resources such as recorded data files, printers, modems, software, and other equipment."

Classification of Networks

According to Dantas (2002), one of the features most used for classifying networks is their geographic coverage. Thus, it is conventionally divided the classification of local networks - LANs (Local Area Networks), metropolitan - MANs (Metropolitan Area Networks) and wide-area - WANs (Wide-Area Networks).

LAN

According to Das ([SD], p 246.) The local area network - LAN "is a fa- mobility communication that provides a high-speed connection between processors, peripherals, communication terminals, and devices in general in a single building or campus.”

LAN is the technology that has a good answer for inter-connecting devices with relatively small distances and with considerable bandwidth.

MAN

Metropolitan networks can be understood as those that provide the interconnection of local area networks in a metropolitan area of a given region.

WAN

When the distances involved in the interconnection of computers are superiors to a metropolitan area and may be geographically dispersed as large as the distance between continents, the correct approach is the geographically distributed network (WAN).

Topologies

Topology can be understood as how communication links and switching devices are interconnected, effectively providing signal transmission between network nodes. [...]

We can say that the physical topology of a local network comprises the physical linkages of the computational elements of the network, while the logical topology of the network refers to how the signal is effectively transmitted between one computer and another.

Bus

In this type of topology, all PCs are physically attached to the same cable, with it, any computer can use it as communication is being made.

Star

The star topology uses a peripheral hub, usually a hub, connecting all machines on the network.

Ring

In this topology, each computer, following a given direction, is connected to the neighbor computer, which in turn tam- well and connected to the neighbor and so on, forming a ring.

Broadcast Media

According to Tanembaum (1997), several physical media can be used for transmission of data. Each has its own niche in terms of bandwidth, delay, cost, and ease of installation and maintenance. The physical resources are grouped into guided media such as copper wire and fiber optics, and unguided media, such as radio waves and laser beams transmitted through the air.

Coaxial Cable

According to Tanembaum (1997), a coaxial cable consisting of a copper wire stretched in the central part, surrounded by an insulating material. The insulation is protected by a cylindrical conductor, usually a strong interknitted loop. The outer conductor is covered by a protective plastic layer.

Twisted Pair

According to Torres (2004), the twisted pair is the most used network cable type currently. There are basically two types of twisted pair: Unshielded, also called UTP (Unshielded Twisted Pair), and shielding, also called STP (Shielded Twisted Pair). The difference between them is precisely the existence in the shielded twisted pair, a mesh around the cable shielding it against electromagnetic interference.

Categories

According to Morimoto (2008a, [nonpaged]), there are cables of category 1 to category 7:

a)    Categories 1 and 2: These two cable categories are no longer recognized by the TIA (Telecommunications Industry Association), which is responsible for defining the wiring patterns. They were used in the past in telephone installations, and category two cables came to be used in Arcnet networks 2.5 megabits and Token Ring 4 megabits but are not suitable for use in Ethernet networks.

b)    Category 3: This was the first pair of wires twisted pattern developed especially for use in networks. The pattern signal is certified for up to 16 MHz, allowing its use in 10BASE-T standard, which is the standard Ethernet network of 10 megabits for cable pair transitional Cado. Still existed a pattern of 100 megabits to Category 3 cable, 100BASE-T4, but it is rarely used and is not supported by all network cards.

c)     Category 4: This category cable has a quality slightly superior and is certified for signal up to 20 MHz. They were used in Token Ring networks of 16 megabits and could also be used in Ethernet networks to replace the category three cables. But in practice, this is unusual. As the categories 1 and 2, the category 4 is no longer recognized by the TIA and cables are no longer manufactured, instead of Category 3 cable, which is still being used in telephone systems.

d)    Category 5: the category five cables are the minimum requirement for 100BASE-TX and 1000BASE-T networks, which are, respectively, network standards 100 and 1000 megabits currently used. The Cat 5 cables follow much stricter manufacturing standards and support frequencies up to 100 MHz, which is a big jump from the cat ropes 3.

e)    Category 6: this category of cable was originally developed for use in Gigabit Ethernet, but with development of the standard cable category five adoptions ended up being delayed because, although the cables category six offer superior quality, the range continues it is only 100 meters, so that, although the best quality cat six cables is always desirable, not just existing Tindo gain much in practice.

f)      There are also cables category seven that may be used in the standard 100 gigabits, which is in the early stages of development.

As the cables category five are sufficient for both networks 100 megabits as 1000, they are the most common and cheaper, but the cables Category 6 and Category 6a are very popular and should replace them over the next few years. The cables are sold originally at 300 meters boxes, or 1000 feet (equivalent to 304.8 meters).

Optical Fiber

According to Torres (2001), "the optical fiber transmits information through light signals instead of electrical signals." The optical fiber is totally immune to noise; therefore, communication is faster.

According to Morimoto (2008c), natural successors of the twisted pair cables are fiber optic cables that support even higher speeds and allow for forward virtually unlimited distances with the use of repeaters. The fiber optic cables are used to create the backbone routers that connect the internet key. Without them, the large network would be much slower and much more expensive access.

According to Das (2002), the optical fibers used in networks are classified according to the way light travels in the cable, these being the monomode and multimode.

Singlemode

In a singlemode class, a single light signal is carried directly in the cable core. The signal can reach distances greater without repetition. This form of light traffic compared with the transmission fiber in the second class (Dantas, 2002).

Multimodal

The multimode fiber is characterized by a light beam that travels along its path, making different refractions in the walls of the cable core (Dantas, 2002).

OSI Model and Model TCP / IP

The OSI model attempts to explain the operation of the network, dividing it into seven layers [...]. Although it is only a theoretical model, which does not need necessarily to be followed to the letter by the network protocols, the OSI model is interesting because it serves as a cue to explain various theoretical aspects of network operation. There are books and courses devoted entirely to the subject, trying to explain everything detailed, sorting everything inside one of the layers, but actually understand the OSI model is not that hard.

The OSI Model

As Torres (2004), to facilitate the interconnection of importers computed systems, the ISO (International Standards Organization) has developed a reference model called OSI (Open Systems Interconnection), so that manufacturers could create protocols from this template.

OSI Model Layers

According to Spurgeon (2000), the OSI reference model is the method to describe how interconnected sets of network hardware and software can be arranged to work concurrently in the networking world. Indeed, the OSI model provides a way to divide the task of the network arbitrarily into separate pieces, which are subject to the formal standardization process.

To do this, the OSI reference model describes seven layers of network functions, described below.

Layer	description
Physicist	This layer takes the frames sent over the link layer and transforms them into signals compatible with the environment where the data should be transmitted.
Data Link	The data link layer takes the received data packet network layer and transforms them into frames that will travel across the network by adding information such as the address of the source network adapter, the destination network adapter address, control data, the data itself and checking cyclic redundancy (CRC).
Network	It is responsible for addressing the packets by converting logical addresses into physical addresses so that the packages to arrive at their destination.
Transport	This layer is responsible for getting data sent by the session layer and divide them into packages to be transmitted to the network layer.
Session	The session layer allows two applications on different computers to establish a sessionCommunication.
Presentation	The presentation layer converts the data format received by the application layer into a common format to be used in the transmission of data.
Application	The application layer is the interface between the communication protocol and the application that requested or receive the information over the network.

The TCP / IP Model

According to Dantas (2002), the reference model is the best-known TCP / IP (Transmission Control Protocol / Internet Protocol). The TCP / IP model was designed in four layers.

Layer	description
network interface (network access)	This layer, network access, is the first TCP / IP model; its function is to support the network layer, attract the physical and logical access services to the physical environment.
	The level inter-network (Internet) is responsible for sending
Inter-network	datagrams from one computer to any
(Internet)	another computer, regardless of their locations
	on the network.
Transport	The transport layer is responsible for providing support to reliably application layer (or not), whether the services offered by the network interface layers and inter-network.
Application	The fourth layer of the TCP / IP is called the application layer. In this layer, the protocols are that support user applications.

Data Communication Protocol

According to Torres (2004), a protocol is the "language" used by devices on a network so that they can understand, that is, exchange information with each other. A protocol is a set of rules governing the communication data (Forouzan, 2006).

Types of Protocols

There are several types of protocols. Next, described are the main ones:

a)    HTTP - HyperText Transfer Protocol - is mainly used for access SAR data on the World Wide Web This protocol allows the transfer of data in the form of simple text, hypertext, audio, video and many others (Forouzan, 2006).

b)    SMTP - Simple Mail Transfer Protocol - This protocol is the default e-mail mechanism internet (Forouzan, 2006);

c)     FTP - File Transfer Protocol - FTP file transfer protocol is the standard mechanism offered by the internet to copy a file from one host to another (Forouzan, 2006);

d)    SNMP - Simple Network Management Protocol - is an Internet management protocol (Dantas, 2002);

e)    DNS - Domain Name Server - this application protocol is fun- to identify IP addresses and maintain a table with the ways of the addresses of some networks on the Internet (Dantas, 2002);

f)      TCP - Transmission Control Protocol - the feature of this protocol is to provide a reliable service between applications (Dantas, 2002);

g)    UDP - User Datagram Protocol - is known for the characteristic of being an optimistic protocol, i.e., it sends all its packages, accredited ing they arrive smoothly and sequentially to the recipient (Dantas, 2002);

h)    IP - Internet Protocol - is the main protocol inter-network level in the TCP / IP architecture (Dantas, 2002);

i)       ICMP - Internet Control Message Protocol - this protocol is to ob- PURPOSE provides control messages in the communication between nodes in a network environment TCP / IP (Dantas, 2002);

j)       ARP - Adress Resolution Protocol - the protocol that maps an IP address in its MAC address (Forouzan, 2006);

k)    RARP - Reverse Resolution Protocol - the protocol that maps a MAC address to an IP address (Forouzan, 2006).

IP Addresses

As Morimoto (2006 [unpaged]), "the IP address is divided into two parts. The first identifies the network to which the computer is the connection, and the second identifies the host within the network. "

Classes Address

According to Morimoto (2006 [unpaged]), to improve the utilization of addresses available, developers - TPC / IP shared the IP address into five classes, called A, B, C, D, and E, and [that] the first three are used for addressing purposes, and the last two are reserved for future expansions. Each class reserves a different number of bytes for addressing the network.

In class A, only the first octet identifies the network; in class, B is used the first two octets, and class C has the first three octets reserved for the network, and only the latter reserved for the identification of hosts within the network.

What differentiates a class of addresses of the other is the value of the first octet. If a number between 1 and 126, have a Class A address A. If the value of the first octet is a number between 128 and 191, then we have a class B address, and finally, if the first octet is a number between 192 and 223, will have a class C address.

Network Active Elements

Hub

According to Torres (2004), the hubs are hubs devices, RESPONSIBLE for centralizing the distribution of data frames in physically connected star networks. Every hub is responsible for replicating repeater, in all its ports, the information received by the network machines.

Switch

According to Torres (2004), switches are bridges that contain multiple ports. They send data frames only to the destination port, unlike the hub, which transmits frames simultaneously to all ports. Thus, the switches can increase network performance.

Router

Routers that are bridges operate at the network layer of the OSI Model. They are responsible for deciding which way to go to interconnect different networks.

Repeater

According to Gallo (2003), the function is to retrieve a signal repeater. Repeaters are also called concentrators and are used in local area networks, increasing its reach.

Bridge

The bridge (bridge)  is an intelligent repeater. It operates on the bed of the link of the OSI model. That means it can read and analyze the data frames that are circulating on the network.

Internet, Intranet, and Extranet

Internet

According to Almeida and Rosa (2000), the internet is a set of interconnected computers networks among themselves, which are scattered all over the world. To - of the services available on the internet are standardized and use the same set of protocols (TCP / IP).

Intranet

According to Wikipedia, an intranet is a private computer network that [it] is based on the suite of Internet protocols. Consequently, all the concepts of the last apply also to an intranet, for example, the client-server paradigm. Briefly, the concept of Intranet can be interpreted as "a private version of the Internet" or a mini-internet confined by an organization.

Extranet

According to Wikipedia, the Extranet of a company is the portion of its computer network that uses the Internet to share part of its information system securely. Taken the term in its broadest sense, the concept is confused with the intranet. An extranet may also be seen as a part of the company that is extended to external users (outside the enterprise network), such as representatives and customers. Another common use of the Extra-net term occurs in the designation of the private part of a site where only registered users can browse previously authenticated by password.

Wireless Networks

A wireless network refers to a computer network without the need to use cables. [...] Their classification is based on the area of coverage: personal or short networks (WPAN), local area networks (WLAN), metropolitan area networks (WMAN), and geographically distributed networks or long-distance (WWAN).

WPAN

Wireless Personal Area Network (WPAN) or personal wireless network, normally [is] used to connect electronic devices physically near you, which you do not want to be detected at a distance (WIKIPEDIA). According to Torres (2004), the main equipment used in this network is Bluetooth and infrared.

Bluetooth

Bluetooth is an open standard for wireless communication, developed by the Bluetooth Special Interest Group - SIG, which includes several companies, including Sony, IBM, Intel, Toshiba, and Nokia.

Unlike Wi-Fi standard, which includes the 802.11b, 802.11a, and 802.11g, used in wireless networks, Bluetooth aims to replace the cables, allowing mobile phones, palmtops, mouses, headsets, etc., exchange data with each other and the PC without cables (MORIMOTO, 2007 [unpaged]).

Infra-Red

The infrared is used in wireless LANs, especially those where you need to connect notebooks.

There are two methods for data transmission using infrared light: direct transmission and diffuse transmission. [...] Indirect transmission, the transmitting and receiving devices have a small opening angle, [so they need to be aligned to transmit the data]. In diffuse transmission, infrared signals are sent in all directions.

WLAN

Wireless LAN or Wireless Local Area Network (WLAN) "is a local network that uses radio waves to make an Internet connection or from a network."

Radio

There are two basic modes of data transmission via radio on- (Figure Below). The non-directional antennas located where the fingertips region of radio waves from the transmitting antenna can capture the transmitted data. [...] This system is widely used in buildings, to connect machines or networks together without cable. The directional transmission, using small satellite dishes, [...] only two networks can communicate. This system has a great advantage, only to transmit data to the receiver (no scattering radio waves to other antennas).

WMAN

Wireless Metropolitan Area Network (WMAN) means metropolitan wireless networks. They enable communication of two nodes distant (MAN) as if they were part of the same local network.

WAN

The Wide Area Network (WAN), wide area network, or long-distance network, also known as the geographically distributed network, is a computer network covering a large geographic area (generally you a country or continent).