Networking Basics

Computer networking is a wide discipline that incorporates numerous computing concepts primarily aimed at improving communication and access to limited (yet sharable) computer resources.

Table of Contents

In this article, we are going to examine the basic concepts of computer networking. Our discussion shall begin by defining a network, after which we will look at network infrastructure, roles of a network administrator, an overview of the different Personal Area Network with a lot of emphasis on LANs and WANs, and an analysis of peer-to-peer networking vs. client-server networking.

To sum it up, we will take a quick glimpse of the various network devices, terminologies, the OSI model, and wrap it up with a brief discussion of collision and broadcast. This chapter is typically a summary of networking fundamentals that prepares us for more of a highly illuminating networking experience in the subsequent chapters.

Networking Basics – What is a switch, router, gateway, subnet, firewall & DMZ

Computer Network: The Meaning

Computer network is a term that refers to any collection of computers that are linked to one another for communication and sharing of data (and information) and other computing resources. Other resources that network also permits network hosts to share applications, data, and a lot more network resources of hardware nature-file servers and printers, among many other devices.

Computer networks may be distinguished according to size, functionality and even location. However, size is the main criterion with which computer networks are classified.

Classification of networks according leads us to the following common types of networks:

LANs
WANs

LAN vs WAN

LAN means Local Area Network, whereas WAN, in full, means Wide Area Network.

LAN

A LAN refers to any group of computers that are linked to one another in a small area like an office or a small building. In a LAN, two or more computers are connected via communication media like coaxial cables, twisted pair copper cables or fiber-optic cables.

It is easy and less costly to set up a LAN since it can do just fine with inexpensive network hardware such as switches, Ethernet cables, and network adapters. The limited traffic allows for faster transmission of data over LANs.

Besides, LANs are easy to manage since they are set up in a small space. Thus, even security enforcement is also enhanced through closer monitoring of activities within the network’s geographical location.

LAN examples include office networks and home-based networks.

Merits of a LAN

LANs have a rather obvious advantage over WANs considering the LANs’ small

geographical coverage, unlike the Wans that stretch over unlimited geographical coverage.

The following are the pros of a LAN:

Ease of installation since it involves a small area within which computers can be connected. The limited area of operation amounts to a limited number of networked machines, which makes it a lot easier to set up a LAN.
Ease of maintenance due to the small network area and few networked computers.
Ease of security enforcement since also due to the relatively small operating environment and a few networked devices

Limitations of a LAN

The limitations of a LAN can be summarized in one sentence by considering its confinement to limited geographical space and the number of networked machines. Thus, it is agreeable to conclude that LANs’ limitation is its inability to accommodate many users, thereby restricting LANs for use within small offices, business settings, learning spaces, and home settings.

WAN

A WAN is a kind of computer network that stretches over large geographical regions-cities, states and even countries. It is bigger than LAN or MAN. It is not restricted to a particular geographical location. It spans over large geographical locations by the use of telephone lines, satellite links or fiber optic cables. The Internet is a perfect example among the existing WANs globally.

WANs are widely embraced for education, government, and business activities.

Examples of WAN

The following examples show just how WANs can connect people limitlessly irrespective of their geographical locations:

Mobile Broadband: 3G or 4G networks are widely serving people in a big region, state or even country.
Private Network: banks create private networks that link different offices established in different locations via a telephone leased line that’s obtained from a telecom company.
Last Mile: telecommunication companies offer internet services to thousands of customers in different cities by simply connecting homes, offices and business premises with fiber.

Notably, the Internet is the most conspicuous example of WANs that connects people in all corners of the universe.

Advantages of WANs

WANs cover large geographical locations reaching out to masses of the human population. The impact of the Internet on people’s lives globally sums up the advantages of a wide area network. Centralized data. WANs support the centralization of data/information. This eliminates a need for individuals to buy back-up servers for their emails and files.
Getting updated files. Programmers get updated files within seconds since software work on live servers.
Quick exchange of messages. WANs use technologies and sophisticated tools that enable a message exchange to happen faster than on most other networks. Communication via Skype and Facebook are two good examples of quick message exchange, thanks to the Internet, one of the popular WANs in the world.
WANs allow the sharing of resources and software. It is possible to share hard drives, RAM, and other resources via wide area networks.
Business without borders. Presently, even people separated by the Pacific can still conduct thriving business without moving an inch from their current location because of the Internet. The world is indeed a global village.
High bandwidth. The use of leased lines for companies increases bandwidth. This, in turn, increases data transfer rates, thereby increasing the productivity of the company.

Disadvantages of WANs

Security issues are escalated as the network size increases. Thus, the issue of insecurity is more concern on a WAN than it is on a LAN or MAN.
High installation cost. Setting a WAN requires the purchase of many costly equipment as well as software applications to manage and administer the network. Routers, switches, and mainframe computers that are needed to serve the network all cost a fortune.
Network troubleshooting is often a big concern since the network spans large geographical locations.

Network Infrastructure

Network infrastructure entails all the necessary resources that lead to the full functionality of the networking concept. That is to say, in other words, that hardware, software, network protocols, human input, and design functions that lead to effective network operation, management and communication; all these do constitute what is conventionally referred to as network infrastructure. In a nutshell, the following are some of the elements of network infrastructure:

Network software
Network hardware
Network protocols
Network services

The hardware aspect of a computer network allows users to physically link network devices as well as an interface to gain access to the resources of a network. Hardware typically includes the physical components that constitute a computer network. The physical components of a network include host machines (computers); connecting devices (routers, hubs and switches); and other peripherals (printers, wireless modems, network cameras, and file servers, among others).

As far as network software is concerned, a Network Operating System (NOS) emerges as number one on the list. However, depending on the nature of a network,

a NOS may not be necessary, particularly in peer-to-peer network arrangement (coming shortly in a different discussion). Besides a NOS, there are numerous software applications that are installed to operate on host machines that network use to perform different tasks on the network.

Network protocols refer to policies and standards that offer details on how network communication takes place. A protocol is a set of conventions, procedures, rules, and regulations that governs the process of network communication. With this in mind, it goes without saying that an understanding of network architecture (which describes the logical arrangement of computers); and the two network models (TCP/IP and OSI) play a crucial role in the comprehension of the entire concept of computer networking.

A computer serves its users with a number of services. The sum total of the network services constitutes the role of the network (network function). Network services include data storage, directory services, email services, file-sharing services, and many more.

In this section, we will discuss peer-to-peer vs. client-server network architectures, the OSI model, network speeds, the role of a network administrator, and collision and broadcast domains.

Peer-to-Peer vs Client-Server

Peer-to-Peer Network Architecture

In this kind of architecture, all computers are connected to one another. All computers have equal privileges and share the responsibility of data processing on equal terms.

This form of computer network architecture is ideal for small computer networks supporting up to 10 computers.

The architecture does not provide for a server role. Special permissions are granted to each computer through an assignment. Unfortunately, issues do arise when the computer with the resource breaks down or malfunctions.

Merits of Peer-To-Peer Networks

The following are the main advantages of Peer-To-Peer Network Architecture:

Less costly since there is no dedicated server.
It’s a small network. Thus, setting up and management of the network is largely easy.
The failure of one machine does not affect the functionality of others. Hence, it is highly reliable.

Demerits of Peer-To-Peer Network Architecture

Peer-to-peer arrangements lack centralized systems. Thus, there is no mechanism for data backup since all data is dissimilar in different locations.
There is no managed security-each computer that has to handle its own security.

Client-Server Network Architecture

In this network model, user computers (known as client computers) rely on a central computer (the server) for resource allocation and enforcement of security. The server handles security, resource, and general network management. On the other, client computers communicate with one another via the central computer/server.

For instance, if the client “A” wishes to send data to client “B,” client A must submit a request for permission to the server. The server then answers back by either granting permission to client A or denying them permission to talk to client “B.” When the server grants client “A” permission to communicate with client “B.” communication can then be initiated by client A to client y instantly or may require to wait for some time.

Pros of Client-Server Network Architecture

Data backup is achievable with the presence of a centralized system.
A dedicated server improves the overall performance through proper organization and management of network resources.
Security enforcement is a notch higher since the central computer administers all shared resources. The speed of resource sharing is higher due to orderly request handling.

The Cons of Client-Server Network Architecture

Dedicated servers are highly expensive. Thus, they render the network quite costly.
The administration of the network must be handled by skilled personnel only. Unlike peer-to-peer networks that do not need any highly skilled individual to administer, client/server networks require qualified personnel for effective administration.

Network Devices

From a physical perspective, a network can be quite simple-just two computers connected together to move data from one to the other over a simple Ethernet cable. That’s not to say, however, that the network will stay simple. For this reason, you should consider every network building block in the initial design even if it is not included in the first phase of implementation.

Even if you are intent on building a home network or small-office network, you ought to anticipate future needs besides those other things intended for purchase and installation without hesitation; either accommodating the need for space, nodes, and wiring right away or building a plan for making the additions and upgrades. Doing so saves time in the long run and may eliminate some frustration when hooking up a new server doesn’t mean that switches, routers, or hubs also have to be changed out.

The following list is a good starting point for identifying the necessary networking components:

Printers
Database hosts
Client workstations and PCs
File servers
Laptops, notebooks, and handhelds
Other peripheral hardware:
- Interface devices
  - Hard drives
  - Network switching and routing components
  - Web cameras
  - Network and end-user software
  - Removable media

Network Speeds

In computer networking, speed and bandwidth are almost interchangeable, but they are not, really. So, what is speed (and bandwidth)?

Whereas network speed is the circuitry bit rate, bandwidth is that “speed,” which ends up being used. Thus, speed refers to the theoretical throughput, while bandwidth is the actual throughput.

In a scenario of the internet, we can define speed in the following ways (bandwidth, actually):

How fast or slow a new connection can be established.
How long it takes to stream a video content comfortably.
How fast or it takes to download content from a website.
How fast or slow it takes a webpage to open.

Bandwidth plays a key role in determining the “speed” of a network. In the world of computer networking, it is not ridiculous to say that bandwidth is a data rate that a network interface (connection).

Ethernet network bandwidth vary immensely from a few megabytes per second (Mbps) to thousands of Mbps. Different Wi-Fi standards define different speeds (bandwidths), as well as other networking technologies.

Numerous factors lead to differences in theoretical and actual network speeds. Some of the factors include:

Network protocols
Communication overheads in the diverse networking hardware components
Operating systems

Importantly, a discussion about network speeds would not be complete without mentioning the term latency. It refers to the time of data transmission from a network host to the server and back. It is measured in milliseconds. It is sometimes considered as a “ping,” which should ideally be recorded at 10ms. High latency is feared to cause slowdowns and buffering.

The OSI Model

OSI is, in full, Open System Interconnection. This model offers a description of the way information and data from a software application is transmitted through a physical media to another software application in a totally unrelated computer. This reference model is made up of seven layers. Each layer has a specific role to play.

The OSI Reference model was born in 1984 by the International Organization (ISO). In modern days, this is taken to be the basic architectural model for inter-computer communication.

In the OSI model, whole tasks are broken down into 7 smaller and manageable chunks. Layers are assigned distinct roles-each layer is assigned a specific task to handle. Also, each layer is sufficiently equipped to handle its tasks independently.

Characteristics of the OSI Model

The OSI model is broadly divided into two layers: upper and lower layers. The upper layers include the following distinct layers:

Transport
Presentation
Application
Session

The lower layers include the following distinct layers:

Physical
Data link
Network

The upper layer of this model primarily handles issues related to applications. Those issues are executed in the software. The closest layer (or the uppermost) to the user is the application layer. The end-user interacts with a software application just as the application software does.

When a layer is said to be an upper layer, it is said so about another. An upper layer is a layer that lies right above the other one.

The lower layer of this model handles issues of data transport. The implementation of the data link, as well as physical layers, occurs in software and hardware. In this model, the physical layer stands as the lowest layer. It is also the nearest to the physical medium. Primarily, the physical layers provide the necessary information to the physical medium.

Roles of Each One of the 7 Layers

We are going to focus on the functions of the unique layers of the OSI Reference model from the lowest to the uppermost.

Physical Layer

Data Transmission: It defines the mode of transmission between two network devices-whether it is full-duplex, half-duplex or simplex mode.
Line Configuration: It offers a clear definition of the way two or more network devices are physically linked.
Signals: the physical layer determines the nature of signals used to transmit information. Topology: The physical layer offers a comprehensive definition of the arrangement of network devices.

Data Link Layer

This layer is charged with the task of ensuring error-free data transfer of data frames over the network. It also defines the data format on the network.

The data link layer ensures that there is reliable and efficient communication between network devices. It is responsible for the unique identification of each device that is found on the network.

The data link layer comprises of the following two layers:

Logical link control layer: It transfers packets to the destination’s network layer. Besides, it identifies the specific address of the network layer of the receiver from the packet header. Furthermore, flow control is implemented in this layer.
Media access control layer: This refers to a link that exists between the physical layer and link control layer. This is what transfers data packets over a network.

The Data Link Layer’s Actual Functions

Framing: the data link layer does the translation of the physical layer’s raw bit stream into data packets referred to as frames. It adds a header and trailer to the data frame. The header contains both receiver and source addresses.
Physical addressing: The physical addressing layer enjoins a header to the frame. This header has the address of the receiver. The frame is transmitted to the receiver whose address is indicated on the header.
Data flow control: This is the data link layer’s primary role. It maintains a constant data rate so that no data is corrupted while in transit.
Error control: This is achieved by the addition of a cyclic redundant check (CRC) on the trailer that is put into the data packet before being sent to the physical layer. In case of any errors, the receiver can request the retransmissions of the corrupted frame.
Access control: This layer determines which of the available network devices is given top priority over the link at a particular moment.

The Network Layer

It is number 3 on the 7 layer OSI Reference model. It handles devices’ IP address assignment and keeps track of device location on the network. Based on network conditions, the layers determines the most favorable path for data transfer from sender to receiver. Another condition that is considered in determining the best path is service priority, among others.

These layers are charged with the responsibility of routing and forwarding packets — routers some of the devices on layer 3. The routers are specified in the network layer and are used to offer routing services in a computer internetwork.

Protocols that are used in the routing of network traffic include IPv6 and IP.

Network Layer Functions

Addressing: This layer ensures that the destination and source addresses are added to the header of the frame. Addressing is helpful in the identification of devices on a network.
Internetworking: The network layer offers a logical link between network devices.
Packetizing: The network layer receives frames from the upper layers and turns them into packets in a process that is conventionally referred to as packetizing. It is realized by the Internet protocol.

The Transport Layer

It is the number 4 layer in the model.

The layer ensures that it follows the order in which they are sent. It makes sure that duplication of data does not occur. This layer’s core business is to ensure that data is transferred totally.

The physical layer receives data from the upper layers and subdivides them further into smaller chunks that are referred to as segments.

The layer provides communication between destination and source —from end-to-end— for data reliability. It can also be termed as end-to-end layer.

There are two protocols that are implemented at this layer:

Transmission control protocol
User datagram protocol

TCP

TCP is a short form of Transmission Control Protocol. It is a standard protocol which allows systems to share messages/information over the internet. The protocol establishes and preserves the link between the hosts.

TCP divides data into smaller units referred to as segments. The resulting segments do not travel over the internet using the same route. They reach the destination in no specific. However, TCP reorders the individual segments at the destination to reconstitute the original message.

User Datagram Protocol (UDP)

It is as well a transport layer protocol. As opposed to TCP, the source does not receive any acknowledgment when the destination receives data. This renders the protocol quite unreliable.

Transport Layer Functions

Whereas the network layer does the transmission of data from one machine to another, it is the transport layer that ensures data transmission to the appropriate processes.

Segmentation and reassembly: This layer receives a message from its upper layer. It then splits the whole message into several small chunks. The layer assigns sequence numbers to each segment for identification. At a destination point, the transport layer reconstitutes the segments based on the sequence numbers to form the original message.
Service-point addressing: Service-point addressing enables computers to run multiple applications simultaneously. It also allows data transmission to the receiver not only from one machine to another machine but also from one process to another process. The transport layer adds a port address or service-point address to the packet.
Flow control: This layer also ensures data control. The data control is done from end to end, but not across one dedicated link.
Connection control: there are two services that the transports offer — connectionless service and connection-based. A connectionless service considers each segment to be a distinct packet. The packets travel through different routes to the destination. On the other hand, the connection-based service makes a connection with the destination machine’s transport for before packets are delivered. In the connection-based service, all packets move on a single route.
Error control: Just like in data control, this is achieved on an end-to-end basis-not across a single link. The transport layer at the source ensures that the message gets to its destination error-free.

The Session Layer

This layer establishes, maintains, and synchronizes the interaction between communicating network devices.

Session Layer Functions

Synchronization: The session layer adds checkpoints in a sequence during data transmission. In case of errors along the way, re transmission of data takes place from the specific checkpoint. The entire process is referred to as synchronization and recovery.
Dialog control: This layer serves as a dialog controller. The layer achieves by initiating dialog between two processes. Alternatively, the layer can be said to authorize communication between one process and another. This can either be half-duplex or full-duplex.

The Presentation Layer

This layer primarily deals with the language and formatting of information that is transferred between two network devices. It is the network’s “translator.”

The presentation layer is a section of the operating system. It is the portion of the operating system that does the conversation of data from a given presentation format to another presentation format.

This layer is also called the Syntax Layer.

Role of the Presentation Layer

The layer does the conversion of data from the sender-based formats into common formats into receiver-dependent formats on the destination computers.

Encryption

The presentation layer performs encryption to ensure the privacy of data. Encryption is the process that involves the conversion of information transmitted from the sender into another unique form that is then transmitted over the network.

Translation

Processes in different systems exchange information as character numbers, character strings, and many more. Different encoding techniques are applied on different computing machines. It is the presentation layer that handles interoperability between them, unlike encoding techniques.

Compression

The presentation compresses data before its transmission. The compression involves the reduction of the number of bits. This process is essential, especially in the transmission of different multimedia like video and audio files.

The Application Layer

This layer offers the interface for users and applications to access resources on the network. It handles network issues like resource allocation, transparency, and many more. This is not an application. It simply plays its application layer role. It provides network services to end-users.

Role of the Application Layer

Access, transfer, and management of files: This layer allows users to access files remotely, retrieve them, and still manage them remotely.
Mail services: This layer offers an email storage and forwarding storage facility.
Directory services: This layer offers the distributed database bases. This is essential in the provision of important information about different objects.

The Network Administrator

For a network to serve its functions as desired, there’s always an individual who’s charged with the responsibility of working tirelessly to make the networking experience an interesting affair. That person, normally operating behind the scenes, is known as the network administrator. A network operator ensures that the network is up-to-date and working properly.

A network administrator performs many tasks as a fulfillment of their mandate. In summary, the following are the main tasks that a network administrator has to perform:

Physical network storage and cloud management.
Basic testing and security enforcement measures.
Offering assistance to network architects with network models design work.
Operating systems and server management.
Software updating and deployment.
Network troubleshooting. Repair work and upgrade of network.
Configuration of network software such as switches, routers and servers.

A network administrator must be highly knowledgeable and skilled in IT, particularly in computer networking. They must be able to think critically and possess strong analytical skills so as to handle complex network issues effectively.

Collision and Broadcast Domains

A collision domain refers to the network portion that is vulnerable to network collisions. Collisions take place when two or more network hosts transmit data packets simultaneously on a single network segment. It must be understood that the efficiency of a network deteriorates when collisions occur. Issues of collisions are rampant networks that rely on hubs for connectivity with host machines and other devices. Hub-based networks are prone to collision issues since ports on a hub are in one collision domain. This is not the case when router-based and switched networks.

A broadcast is forwarded in a broadcast. Thus, a broadcast refers to the network segment where a broadcast is relayed.

A broadcast domain is composed of all network devices that communicate at the data link layer via a broadcast. By default, switch and hub ports belong to the same domain. On the contrary, router ports belong to different domains. Also, a router cannot forward a broadcast from a broadcast domain to another.