Networking, Internet and Distributed Computing Fundamentals

Introduction

Modern organizations depend on connected computers rather than isolated machines. A retail transaction may be processed by a point-of-sale terminal, an application server, a payment gateway, and a cloud database. Employees collaborate through internet services, customers access web applications through mobile devices, and managers use distributed analytics platforms to support decisions. Networking therefore forms an essential foundation for business information systems.

A computer network is a collection of interconnected devices that exchange data and share resources through communication media and agreed rules. The devices may include computers, servers, printers, smartphones, sensors, and network equipment. Networking enables communication, resource sharing, centralized management, remote access, electronic commerce, and distributed computing.

This chapter explains network classifications, topologies, devices, protocols, addressing, client–server and peer-to-peer models, internet and web services, connectivity options, virtualization, cloud computing, and edge computing.

Learning Objectives

After studying this chapter, a student should be able to:

1. Classify networks by geographical scope and purpose.
2. Explain common network topologies and their consequences.
3. Identify major networking devices and their roles.
4. Describe protocols, layered communication, and addressing.
5. Compare client–server and peer-to-peer architectures.
6. Explain the internet, the World Wide Web, and common internet services.
7. Evaluate connectivity alternatives for organizational use.
8. Explain virtualization, cloud-computing service models, and edge computing.
9. Assess network choices in terms of performance, cost, security, reliability, and scalability.

1. Foundations of Data Communication and Networking

Data communication is the transfer of digital information between two or more devices. A basic communication system contains a sender, receiver, message, transmission medium, and protocol. The message may be text, audio, video, a transaction, or control information. The medium may be copper cable, optical fibre, radio, or satellite transmission.

Network performance is commonly discussed using several measures:

• Bandwidth is the maximum data-carrying capacity of a link, normally expressed in bits per second.
• Throughput is the actual rate achieved after overhead, congestion, errors, and device limitations are considered.
• Latency is the time required for data to travel from source to destination. It is usually measured in milliseconds.
• Jitter is variation in packet delay. High jitter can disrupt voice and video.
• Packet loss occurs when transmitted packets fail to reach their destination.
• Availability is the proportion of time a service remains usable.

A useful approximation for a file-transfer time, when the link is not congested, is:

The formula assumes that file size, throughput, and protocol overhead are appropriately measured. For example, a 100 MB decimal file contains approximately 800 megabits. At an effective throughput of 80 Mbit/s, the ideal transfer time is 10 seconds. Real transfer time may be longer because of encryption, acknowledgements, congestion, and retransmissions.

Networks divide information into packets. Each packet contains control information, such as source and destination addresses, and a portion of the user data. Packet switching allows many users to share communication links efficiently. Unlike circuit switching, which reserves a dedicated path for the duration of a communication session, packet switching dynamically forwards packets through available routes.

2. Types of Networks

Networks can be classified by geographic scope, ownership, technology, or purpose.

A PAN normally uses short-range technologies such as Bluetooth. A LAN is usually owned and managed by one organization and offers relatively high speed and low latency. A WAN connects geographically separated networks through leased lines, internet connections, microwave links, or cellular services. The internet is a global interconnection of independently administered networks, not one centrally owned network.

A VPN creates a logical private connection across a public or shared infrastructure. Encryption and authentication protect traffic, but a VPN does not automatically make an endpoint secure. Weak passwords, compromised devices, or incorrectly configured services can still create risks.

3. Network Topologies

Topology describes how devices and communication links are arranged. Physical topology concerns actual cabling or radio coverage; logical topology concerns how data flows.

Bus topology

In a bus topology, devices share a common backbone cable. It is inexpensive for small installations but a damaged backbone can affect the entire network. Shared access also creates potential contention. Modern business Ethernet networks rarely use the traditional physical bus arrangement.

Star topology

In a star topology, each device connects to a central switch or access point. It is easy to install, diagnose, and expand. Failure of one cable usually affects only one device, but failure of the central device can disrupt many users.

Ring topology

In a ring, each device connects to two neighbouring devices, forming a loop. Data may travel around the ring in one or both directions. A ring can provide predictable access, but a break may interrupt communication unless protection mechanisms or a redundant ring are provided.

Mesh topology

In a mesh, devices have multiple connections. A full mesh connects every device to every other device; a partial mesh provides selected redundant paths. Mesh designs improve resilience but increase installation and management costs. Wireless mesh networks are used where cabling is difficult.

Tree and hybrid topologies

A tree combines multiple star networks in a hierarchy. It is common in organizations because access switches can connect to distribution and core layers. A hybrid topology combines two or more designs. For example, an enterprise may use star topology inside buildings and redundant WAN links between sites.

Figure: Star, ring, and partial-mesh network topologies.

Topology selection is a business decision as well as a technical decision. A small office may prioritize low cost and easy installation, while a hospital or financial institution may prioritize redundancy and continuity of service.

4. Networking Devices

Networking devices operate at different levels of communication and perform different functions.

• A Network Interface Card (NIC) connects a device to a network and usually has a hardware address called a MAC address.
• A hub repeats incoming signals to all connected ports. It is largely obsolete because it creates shared traffic and does not intelligently select recipients.
• A switch forwards frames within a LAN using MAC addresses. Managed switches support features such as VLANs, monitoring, access control, and link aggregation.
• A router connects different networks and forwards packets using logical IP addresses and routing tables.
• A wireless access point connects wireless devices to a wired or wireless network. It manages radio communication and security associations.
• A modem converts signals for transmission over a particular access technology, such as cable or DSL. In contemporary equipment, modem, router, switch, and wireless functions may be integrated.
• A gateway connects systems using different protocols or architectural environments. In everyday usage, the default gateway is commonly the router used to reach other networks.
• A firewall permits or blocks traffic according to security rules. It may be implemented in hardware, software, or a cloud service.
• A proxy receives requests on behalf of clients. It can provide filtering, caching, access control, or privacy-related functions.
• A load balancer distributes requests among servers to improve performance and availability.

A switch generally operates within a local network, whereas a router makes forwarding decisions between networks. These roles can coexist in one physical device, but the conceptual distinction remains important.

5. Protocols and Layered Communication

A protocol is a set of rules governing communication. It defines message formats, addressing, timing, error handling, and the actions taken when communication succeeds or fails. Without protocols, devices from different manufacturers could not reliably interoperate.

Layering divides communication into manageable functions. The OSI reference model describes seven conceptual layers: physical, data link, network, transport, session, presentation, and application. The practical internet architecture is often represented using four or five layers: link, internet, transport, and application, with physical transmission sometimes shown separately.

When an application sends data, each layer adds control information, a process called encapsulation. At the destination, the layers remove and interpret that information through decapsulation.

Important protocols include:

• Ethernet: common wired LAN technology.
• Wi-Fi: wireless LAN technology specified by IEEE 802.11 standards.
• IP: provides logical addressing and packet delivery across interconnected networks.
• TCP: provides reliable, ordered, connection-oriented delivery using acknowledgements and retransmission.
• UDP: provides lightweight datagram delivery without TCP’s built-in reliability mechanisms. It is useful where low delay or application-controlled reliability is preferred.
• HTTP and HTTPS: support transfer of web resources; HTTPS protects HTTP using encryption and authentication through TLS.
• DNS: translates human-readable domain names into IP addresses and other records.
• SMTP: transfers email between mail servers; other protocols such as IMAP support mailbox access and synchronization.
• DHCP: automatically supplies network configuration such as an IP address, subnet information, gateway, and DNS server.
• SSH: provides authenticated and encrypted remote administration.

TCP and UDP are not competing universal replacements. TCP is appropriate when complete, ordered delivery is essential. UDP can be appropriate for real-time applications, DNS queries, streaming protocols, and applications that implement their own recovery or tolerate loss.

6. Addressing and Naming

A network needs addresses to identify communication endpoints. Different layers use different forms of addressing.

A MAC address identifies a network interface at the data-link level. It is used primarily for local delivery. An IP address identifies an interface within an internetworking scheme. A port number identifies a service or process at the transport layer. A domain name provides a human-friendly label that DNS can resolve to an IP address.

IPv4 and IPv6

IPv4 uses 32-bit addresses, commonly written in dotted-decimal form, such as 192.168.1.20. IPv6 uses 128-bit addresses, written in hexadecimal groups, such as 2001:db8::20. IPv6 provides a vastly larger address space and supports modern addressing and autoconfiguration features. IPv4 remains widely deployed, so many organizations use dual-stack operation or translation mechanisms.

A subnet divides an IP network into logical portions. In IPv4, a prefix length indicates how many bits identify the network. For example, a /24 prefix contains 24 network bits and 8 host bits. Under the conventional simplified calculation, the number of addresses is:

where h is the number of host bits. A /24 therefore has 2⁸ = 256 total addresses. In a traditional subnet, two addresses are reserved for the network identifier and broadcast address, leaving 254 assignable host addresses. This exception does not apply identically to every modern addressing context, so the addressing design must be stated before calculating usable hosts.

Private IPv4 addresses are used inside organizations and are not directly routed across the public internet. A router commonly uses Network Address Translation (NAT) to allow multiple private devices to share a public address. NAT conserves IPv4 addresses but can complicate inbound connections, logging, and some peer-to-peer applications.

DNS separates naming from addressing. A user can access a service through a name even if the underlying server address changes. DNS records can identify addresses, mail servers, aliases, and other service information.

7. Client–Server and Peer-to-Peer Models

In the client–server model, clients request services and servers provide them. A web browser is a client, while a web server delivers pages or application responses. A database server stores and processes data for application clients. Centralization simplifies administration, security policy, backup, and auditing, but servers may become performance bottlenecks or single points of failure unless redundancy is designed.

A multi-tier application commonly separates:

1. Presentation, such as a browser or mobile interface.
2. Application logic, such as transaction validation.
3. Data management, such as a database system.

In a peer-to-peer (P2P) model, participating devices can act as both clients and servers. Resources are distributed among peers rather than controlled by one central server. P2P can improve scalability and reduce dependence on a central system, but it creates challenges in trust, coordination, security, data consistency, and support.

Most real systems are hybrid. A collaboration platform may use centralized identity and billing but distribute content among peers or edge caches. The appropriate model depends on control requirements, data sensitivity, availability, traffic patterns, and operating cost.

8. The Internet and Web Services

The internet is a worldwide system of interconnected networks that communicate primarily through the Internet Protocol suite. It is decentralized: universities, companies, governments, carriers, and other organizations operate networks and exchange traffic through agreements and technical standards.

The World Wide Web is an internet-based information system built mainly on URLs, HTTP or HTTPS, web browsers, web servers, and linked resources. The internet and the web are not synonymous. The web is one service that uses the internet; email, file transfer, voice communication, remote administration, and domain-name resolution are other internet services.

Common web services include:

• Websites and web applications, which deliver information or interactive functions through browsers.
• Search engines, which discover, index, rank, and retrieve information.
• Email, which supports asynchronous communication and business records.
• File transfer and storage, which support collaboration and distribution.
• Voice and video conferencing, which provide real-time communication.
• Streaming, which delivers audio or video progressively rather than requiring the entire file first.
• Application Programming Interfaces (APIs), which allow systems to exchange structured data and invoke functions.
• Electronic commerce, which combines web interfaces, payment services, inventory systems, logistics, and customer data.

HTTPS is essential for protecting confidentiality and integrity in transit and for helping users authenticate the intended service. It does not guarantee that a website is honest, that its database is secure, or that a user’s device is free from malware. Secure design also requires authentication, authorization, input validation, patching, monitoring, and data protection.

9. Connectivity Choices and Network Evaluation

Organizations select connectivity by balancing performance, cost, geographic reach, reliability, security, and manageability.

A connectivity assessment should identify users, applications, traffic volume, peak demand, latency sensitivity, availability requirements, compliance obligations, and recovery needs. A video-conferencing service is sensitive to latency, jitter, and packet loss; a backup service may prioritize capacity and price.

A simple availability calculation is:

Both uptime and downtime must be measured over the same period. If a service operates for 8,760 hours in a year and is unavailable for 8.76 hours, its availability is approximately 99.9%. This percentage should not be interpreted without considering maintenance exclusions, service-level definitions, and the business impact of an outage.

10. Virtualization and Cloud Computing

Virtualization abstracts physical computing resources so that multiple logical environments can share one physical machine. A hypervisor creates and manages virtual machines (VMs). Each VM can run an operating system and applications as though it had dedicated hardware. Virtualization improves utilization, isolation, provisioning speed, testing, and disaster recovery.

A Type 1 hypervisor runs directly on physical hardware, while a Type 2 hypervisor runs on top of a host operating system. Containers provide a lighter form of isolation by sharing the host operating system kernel. Virtual machines are often preferred when stronger operating-system separation is required; containers are efficient for packaging and deploying compatible applications.

Virtualization does not eliminate resource constraints. CPU, memory, storage, and network resources remain finite. Excessive consolidation can create contention, and a compromised hypervisor or poorly isolated environment can affect many workloads. Management must therefore include capacity planning, patching, access control, monitoring, and backup.

Cloud computing provides on-demand access to shared configurable resources, often with rapid provisioning and usage-based charging. Important service models are:

• Infrastructure as a Service (IaaS): virtual machines, storage, and networks are provided while the customer manages operating systems and applications.
• Platform as a Service (PaaS): the provider manages more of the runtime platform, allowing developers to focus on application code and data.
• Software as a Service (SaaS): a complete application is delivered to users, usually through a browser or client program.

Deployment models include public cloud, private cloud, hybrid cloud, and multi-cloud. Cloud benefits include elasticity, global reach, managed services, and reduced capital expenditure. Risks include vendor lock-in, unclear data location, service outages, unexpected usage charges, compliance issues, and shared-responsibility misunderstandings.

Under the shared-responsibility model, the provider secures the underlying cloud infrastructure, while the customer remains responsible for aspects such as identities, configurations, data, and applications according to the service model. Moving to the cloud is not itself a security strategy.

11. Edge Computing and Distributed Computing

Distributed computing divides computation and data processing across multiple networked computers that cooperate to provide a service. Distributed systems may improve scalability, resilience, geographic reach, and performance, but they introduce coordination problems. Network delays, partial failures, inconsistent data, clock differences, and security boundaries must be managed.

Edge computing places computation closer to users, devices, or data sources rather than sending every task to a distant data centre. An industrial sensor may send data to an on-site gateway that detects an unsafe condition immediately, while summarized data is later sent to the cloud for long-term analysis.

Edge computing is valuable when applications require low latency, reduced bandwidth use, local operation during intermittent connectivity, or data-locality control. It also creates additional devices to secure and manage. Edge nodes may have limited processing power, inconsistent physical protection, and difficult update conditions.

Cloud, edge, and local computing are not mutually exclusive. A good architecture assigns each task to the most appropriate location:

• Immediate control decisions may occur at the edge.
• Regional aggregation may occur at a local or regional platform.
• Large-scale analysis and model training may occur in the cloud.
• Sensitive records may remain within a controlled organizational environment.

Artificial intelligence is increasingly distributed in this way. A device may perform inference locally for rapid response, while a cloud service trains models using aggregated and governed data. Designers must address model updates, privacy, bias, explainability, and the consequences of incorrect automated decisions.

12. Applied Case: Retail Branch Connectivity

Background

Consider an illustrative mid-sized retailer with 40 stores. Each store uses point-of-sale terminals, inventory scanners, staff Wi-Fi, security cameras, and a cloud-based inventory application.

Problem

Some stores experience slow transactions during peak periods. A few locations lose connectivity during storms. Management is considering a more expensive leased-line network, but the finance department wants to control recurring costs.

Analysis

Point-of-sale transactions require reliable connectivity and secure access to the central application. Cameras generate substantial continuous traffic but do not always require the same low latency as payment authorization. Staff Wi-Fi should be separated from payment systems. A single internet connection at each store creates a local single point of failure.

Approach and solution

The retailer adopts a segmented design using VLANs or equivalent logical separation for payment, business operations, cameras, and guest access. Each store receives two different connectivity paths where practical: primary fibre or cable broadband and secondary cellular connectivity. A secure VPN links stores to central services. Traffic prioritization gives transaction systems appropriate priority, while cameras use local recording and upload selected footage during less congested periods.

Cloud monitoring reports link availability, latency, packet loss, and unusual traffic. Identity-based access, multifactor authentication, encrypted connections, regular patching, and tested recovery procedures complement the network design.

Outcome

The approach improves resilience without requiring a dedicated leased line to every store. Costs are controlled through selective use of premium connectivity at high-volume locations. The retailer also gains better visibility into incidents and separates business-critical traffic from less urgent traffic.

Learning points

1. Network design must begin with business requirements rather than equipment selection.
2. Redundancy is valuable only when failover is tested.
3. Segmentation limits the effect of compromise and congestion.
4. Performance should be measured using application-relevant metrics.
5. A hybrid connectivity strategy can be more economical than a single universal solution.

13. Governance, Security, and Ethical Considerations

Networking decisions affect privacy, safety, equity, and organizational accountability. Organizations should collect only necessary data, define retention periods, and protect personal information in transit and at rest. Monitoring must have a legitimate purpose, transparent policy, and appropriate access controls.

Important governance practices include asset inventories, documented network diagrams, change management, least-privilege access, strong authentication, encryption, patch management, vulnerability assessment, incident response, logging, backup, and disaster recovery testing. Network segmentation and zero-trust principles can reduce the assumption that everything inside a corporate network is automatically trustworthy.

Connectivity choices may also create a digital divide. A cloud-first or video-intensive business process may exclude employees or customers with limited bandwidth. Accessibility, regional availability, affordability, and disaster conditions should therefore be considered in service design.

14. Contemporary Developments and Best Practices

Current network environments increasingly combine software-defined networking, wireless 5G, IPv6, cloud-native applications, containers, content-delivery networks, and automated monitoring. These developments improve flexibility but increase architectural complexity.

Best practices include:

• Design for security from the beginning rather than adding it later.
• Use documented addressing plans and consistent naming.
• Separate critical, administrative, guest, and device traffic.
• Measure latency, throughput, jitter, loss, and availability continuously.
• Provide redundancy for business-critical links and equipment.
• Test backups, failover, and incident-response procedures.
• Apply the principle of least privilege to users, services, and administrators.
• Evaluate total cost of ownership, including migration, training, licensing, data transfer, and exit costs.
• Select cloud and edge locations according to latency, privacy, resilience, and regulatory needs.
• Review contracts for service levels, data ownership, portability, and breach notification.

Chapter Summary

A computer network connects devices so that they can exchange data and share services. Networks are classified as PANs, LANs, WLANs, MANs, WANs, SANs, and VPNs according to scope and purpose. Bus, star, ring, mesh, tree, and hybrid topologies provide different balances of cost, manageability, redundancy, and failure tolerance.

Switches connect devices within local networks, routers connect different networks, access points provide wireless access, firewalls enforce traffic rules, and gateways connect differing environments. Protocols define communication rules. The internet protocol suite uses IP for internetworking and transport protocols such as TCP and UDP for application communication. MAC addresses, IP addresses, ports, and domain names identify different aspects of network communication.

Client–server systems centralize services and administration, while peer-to-peer systems distribute resources among participating devices. The internet supports the web, email, file transfer, conferencing, streaming, APIs, and electronic commerce. Connectivity must be assessed using business requirements, cost, capacity, latency, availability, security, and geographic conditions.

Virtualization abstracts physical resources into virtual machines or containers. Cloud computing offers IaaS, PaaS, and SaaS through on-demand resources, while edge computing places processing near users or data sources. Distributed architectures can improve scalability and responsiveness but require careful management of failure, consistency, security, privacy, and governance.

Key Terms

Network: Connected devices and communication links that exchange data.

LAN: A network covering a limited local area such as a building or campus.

WAN: A network connecting geographically separated locations.

Topology: The physical or logical arrangement of network components.

Switch: A device that forwards local-network frames using MAC addresses.

Router: A device that forwards packets between different networks.

Protocol: Rules governing the format and exchange of data.

Packet: A formatted unit of data transmitted across a packet-switched network.

IP address: A logical address used to identify a network interface.

MAC address: A link-layer identifier associated with a network interface.

DNS: A naming service that maps domain names to network information.

TCP: A reliable, ordered transport protocol.

UDP: A lightweight datagram transport protocol without built-in delivery guarantees.

Client–server model: An architecture in which clients request services from servers.

Peer-to-peer model: An architecture in which participating devices can provide and consume services.

Virtualization: Abstraction of physical computing resources into logical environments.

Cloud computing: On-demand access to shared computing resources over a network.

Edge computing: Processing performed near users or data sources.

VPN: An encrypted logical connection across a shared or public network.

Availability: The proportion of time that a service is operational and accessible.

Review Questions

Short Answer Questions

1. Distinguish between a LAN and a WAN.
2. What is the principal function of a router?
3. Explain the difference between bandwidth and throughput.
4. What role does DNS play on the internet?
5. Define edge computing.

Long Answer Questions

1. Compare star, ring, bus, and mesh topologies, including their advantages and limitations.
2. Explain the roles of Ethernet, IP, TCP, UDP, HTTP, HTTPS, DNS, and DHCP.
3. Compare client–server and peer-to-peer architectures in terms of control, scalability, security, and reliability.
4. Explain IaaS, PaaS, and SaaS and discuss the shared-responsibility model.
5. Discuss the factors an organization should consider when selecting connectivity for branch offices.

Analytical / Application Questions

1. A company has offices in three cities, cloud-based applications, guest Wi-Fi, and sensitive financial systems. Propose a network architecture and justify your choices of topology, segmentation, connectivity, and security controls.
2. A manufacturing plant must respond to equipment abnormalities within milliseconds, but it also wants long-term analytics. Decide which functions should be placed at the edge and which should be placed in the cloud. Explain the trade-offs.
3. A retailer is choosing between one expensive leased line and two less expensive internet connections with VPN failover. Analyze the decision using availability, cost, performance, security, and operational-management criteria.

Multiple Choice Questions with Answers

Question 1

Which device primarily forwards frames within a local network using MAC addresses?

A. Router

B. Switch

C. Modem

D. Firewall

Correct Answer: B. Switch

Question 2

Which topology normally connects each device to a central switch?

A. Bus

B. Ring

C. Star

D. Full mesh

Correct Answer: C. Star

Question 3

What is the main purpose of DNS?

A. Encrypting all network traffic

B. Translating domain names into network information

C. Assigning processor time to applications

D. Compressing video files

Correct Answer: B. Translating domain names into network information

Question 4

Which protocol is designed to provide reliable, ordered transport?

A. UDP

B. IP

C. TCP

D. DNS

Correct Answer: C. TCP

Question 5

A subnet has an IPv4 prefix of /24. Under the conventional calculation, how many total addresses does it contain?

A. 24

B. 128

C. 254

D. 256

Correct Answer: D. 256

Question 6

Which statement best describes a peer-to-peer network?

A. Only one central server can provide resources

B. Participating devices can both request and provide services

C. It cannot use encryption

D. It is limited to wireless connections

Correct Answer: B. Participating devices can both request and provide services

Question 7

Which cloud service model gives customers virtual machines, storage, and networks while leaving operating-system management largely to the customer?

A. SaaS

B. PaaS

C. IaaS

D. PAN

Correct Answer: C. IaaS

Question 8

What is a major reason for using edge computing?

A. To increase physical distance from data sources

B. To reduce latency by processing data near its source

C. To eliminate all security requirements

D. To replace every form of cloud computing

Correct Answer: B. To reduce latency by processing data near its source

Question 9

A service operates for 8,760 hours and is unavailable for 8.76 hours. Its approximate availability is:

A. 90.0%

B. 95.0%

C. 99.0%

D. 99.9%

Correct Answer: D. 99.9%

Question 10

Which statement about HTTPS is most accurate?

A. It guarantees that every website is trustworthy

B. It protects HTTP communication using TLS but does not remove all application risks

C. It is used only for email transfer

D. It eliminates the need for user authentication

Correct Answer: B. It protects HTTP communication using TLS but does not remove all application risks