Networking, Internet and Cloud Computing

Introduction

Modern organisations depend on digital connectivity. Employees collaborate across offices, customers access services through mobile applications, suppliers exchange information electronically, and managers use cloud platforms to analyse operations. These activities are enabled by computer networks, the Internet, wireless communication, virtualisation and cloud computing.

For a business student, networking is not only a technical subject. It affects service quality, cybersecurity, cost, business continuity, outsourcing and organisational strategy. A manager must understand what different technologies do, how they are governed, and which sourcing decision best fits a firm’s objectives.

Learning Objectives

After studying this chapter, you should be able to:

1. Describe major network types and their business uses.
2. Explain essential networking protocols and the architecture of the Internet.
3. Distinguish wired and wireless networking systems.
4. Explain cloud service and deployment models.
5. Describe virtualisation and edge computing.
6. Evaluate managerial, security, governance and technology-sourcing considerations.

1. Foundations of Computer Networking

A computer network is a collection of connected devices that exchange data and share resources according to agreed rules. Devices may include computers, servers, smartphones, printers, sensors, routers and data-centre systems. A network provides three broad capabilities:

• Communication: exchanging messages, files, voice and video.
• Resource sharing: sharing applications, storage, printers and Internet access.
• Coordination: supporting distributed business processes such as inventory, payment and customer service.

A network is commonly analysed through its nodes, links and protocols. A node is a connected device; a link is a communication path; and a protocol is a set of rules governing communication. Network performance is often discussed using bandwidth, latency, throughput, availability and packet loss.

Bandwidth is the theoretical data-carrying capacity of a link, measured in bits per second. Throughput is the actual rate achieved. Latency is the time required for data to travel between endpoints. A high-bandwidth connection may still perform poorly for interactive applications if latency is high. Availability refers to the proportion of time a service is operational.

Network Types by Geographic Scope

A personal area network (PAN) connects devices close to an individual. A local area network (LAN) is usually owned and administered by one organisation. Ethernet is common in wired LANs, while Wi-Fi is common in WLANs. A wide area network (WAN) connects geographically separated LANs using carrier services, leased links, virtual private networks or the public Internet.

Networks may also be classified by organisational relationship. In a client–server network, client devices request services from dedicated servers. This supports centralised security and administration. In a peer-to-peer network, devices can act as both clients and servers. Peer-to-peer arrangements may be economical for small groups but are generally harder to govern at scale.

Network Topologies

A topology describes how devices are arranged. In a star topology, devices connect to a central switch. It is easy to manage, and failure of one cable usually affects only one device, but failure of the central device can disrupt the network. In a bus topology, devices share one backbone cable; it is simple but less resilient. In a ring topology, each device connects to two neighbours. In a mesh topology, devices have multiple interconnections, providing resilience but increasing cost and complexity. Modern enterprise networks commonly use hierarchical star arrangements, while the Internet itself is a highly interconnected collection of autonomous networks rather than one simple topology.

2. Protocols and Network Architecture

A protocol specifies how data is formatted, addressed, transmitted, received and interpreted. Without protocols, devices manufactured by different organisations could not communicate reliably.

Layered Models

The OSI model is a conceptual seven-layer framework:

1. Physical
2. Data link
3. Network
4. Transport
5. Session
6. Presentation
7. Application

The Internet commonly uses the TCP/IP model, usually represented by four or five layers: link, Internet, transport and application, with the physical function sometimes treated separately. Layering reduces complexity because each layer performs a defined function and provides services to the layer above it.

At the sending device, application data is encapsulated as it moves down the layers. Headers containing control information are added. At the receiving device, the headers are removed as data moves upward. This process allows an application such as a browser to operate without directly controlling electrical signals or radio transmission.

Important Protocols

• Ethernet: A major standard for wired LAN communication. It defines framing and media-access behaviour.
• Wi-Fi (IEEE 802.11): A family of wireless LAN standards using radio communication.
• Internet Protocol (IP): Provides logical addressing and routing of packets between networks. IPv4 uses 32-bit addresses; IPv6 uses 128-bit addresses and provides a vastly larger address space.
• Transmission Control Protocol (TCP): Establishes reliable, ordered delivery using acknowledgements, sequencing and retransmission. It is suitable when completeness matters, such as file transfer.
• User Datagram Protocol (UDP): Provides a lightweight, connectionless transport service. It does not guarantee delivery or ordering, but its low overhead is useful for real-time voice, video and some name-resolution functions.
• Domain Name System (DNS): Translates human-readable domain names into IP addresses. It is a distributed hierarchical naming system.
• Hypertext Transfer Protocol (HTTP): Supports communication between web clients and servers. HTTPS is HTTP protected by cryptographic protocols, typically Transport Layer Security (TLS).
• Simple Mail Transfer Protocol (SMTP): Used primarily for sending email between mail systems. Other protocols, such as IMAP, support mailbox access and synchronisation.
• Dynamic Host Configuration Protocol (DHCP): Automatically supplies devices with configuration such as an IP address, gateway and DNS server.
• Secure Shell (SSH): Provides authenticated and encrypted remote administration.

Addressing, Routing and Switching

A MAC address identifies a network interface at the local data-link level, whereas an IP address identifies a logical endpoint used for routing across networks. A switch forwards frames within a LAN, usually using MAC-address information. A router connects different networks and forwards packets according to routing information.

Routing is the process of selecting a path for packets. Routers use routing tables, which may be configured manually or learned through routing protocols. The Internet is not controlled by one central network operator. It consists of thousands of autonomous systems—networks under a common administrative authority—interconnected through commercial, governmental, academic and private arrangements. The Border Gateway Protocol helps autonomous systems exchange reachability information.

3. Internet Architecture and Services

The Internet is a global network of networks based principally on the TCP/IP protocol suite. A user’s device typically connects through an access network to an Internet service provider. The provider connects to other networks through regional, national and international links. Internet exchange points allow networks to exchange traffic directly, often improving efficiency.

The Internet is distinct from the World Wide Web. The Internet is the underlying global communication infrastructure; the Web is a service operating on it using browsers, web servers, URLs, HTTP and related technologies. Email, file transfer, streaming and online gaming are other Internet services.

When a user enters a web address, the browser generally performs several steps:

1. DNS resolves the domain name to an IP address.
2. The device establishes transport communication, commonly using TCP or a modern secure transport protocol.
3. TLS may authenticate the server and establish encryption for HTTPS.
4. The browser sends an HTTP request.
5. Servers return data, which the browser interprets and displays.

Content delivery networks (CDNs) place copies of content near users, reducing latency and congestion. Internet applications also commonly use application programming interfaces (APIs), which allow software systems to exchange structured requests and responses. APIs support integration between, for example, a retailer’s website, payment gateway, logistics provider and cloud analytics platform.

The Internet is best understood as a distributed and layered architecture. This creates scalability and resilience, but it also creates dependencies. A business may depend on an Internet service provider, DNS provider, cloud platform, certificate authority, CDN and several software vendors. Concentration in any one provider can create operational and strategic risk.

4. Wired and Wireless Networking Systems

Wired Networks

Wired networks use media such as copper twisted-pair cable, coaxial cable or optical fibre. Copper Ethernet is common inside buildings because it is relatively inexpensive and easy to install. Fibre transmits data using light, supports high capacity and long distances, and is less vulnerable to electromagnetic interference. It is therefore important in data centres, telecommunications backbones and inter-building connections.

Wired connections generally offer predictable performance, strong physical control and lower exposure to radio interference. Their disadvantages include installation cost, limited mobility and the practical difficulty of connecting temporary or moving devices.

Wireless Networks

Wireless systems transmit through electromagnetic waves. Wi-Fi connects devices to access points, which connect them to a wired or wireless network. Wi-Fi is convenient but performance depends on distance, obstacles, radio interference, channel use, device density and access-point capacity.

Cellular networks use geographically distributed cells served by base stations. A mobile device changes its serving cell as it moves, a process known as handover. Generations of cellular technology have progressively improved capacity, latency and support for connected devices. Fifth-generation systems are designed for enhanced mobile broadband, large-scale machine communication and some low-latency applications, although actual performance depends on deployment, spectrum and network conditions.

Short-range technologies such as Bluetooth support peripherals and personal devices. Near-field communication supports very short-range interactions such as contactless payment. Low-power wide-area technologies support sensors that transmit small quantities of data over long distances while conserving battery power.

Wireless security requires strong authentication and encryption, careful access-point configuration, network segmentation and monitoring. Public or poorly configured wireless networks can enable eavesdropping, unauthorised access and rogue access-point attacks. A business should separate guest traffic from internal systems and avoid treating wireless convenience as a substitute for security governance.

5. Cloud Computing

Cloud computing is a model for providing shared, configurable computing resources—such as servers, storage, networks, databases and applications—over a network, usually with rapid provisioning and usage-based measurement. The National Institute of Standards and Technology identifies five widely used characteristics:

1. On-demand self-service.
2. Broad network access.
3. Resource pooling.
4. Rapid elasticity.
5. Measured service.

Resource pooling means that a provider serves multiple customers using shared physical infrastructure while logically separating their environments. Elasticity means capacity can expand or contract as demand changes. Measured service supports monitoring and billing according to consumption or subscription.

Cloud Service Models

These models represent different divisions of responsibility. IaaS provides flexibility but requires more technical administration. PaaS reduces infrastructure work and can accelerate development, though it may increase dependence on the provider’s tools. SaaS can be adopted quickly, but the organisation has less control over architecture and release timing.

The shared-responsibility model is central to cloud governance. The provider is responsible for aspects such as physical facilities and, depending on the service, the underlying infrastructure. The customer remains responsible for appropriate identity management, data classification, configuration, access permissions and legal use. The exact boundary differs by provider and service; it must never be assumed.

Cloud Deployment Models

• Public cloud: Infrastructure is operated by a provider and shared among customers through logical isolation. It offers scalability and rapid access without owning data-centre hardware.
• Private cloud: Cloud-like resources are dedicated to one organisation. It may provide greater control or support particular regulatory requirements, but the organisation bears more cost and management responsibility.
• Hybrid cloud: Combines private resources with public cloud services, allowing selected workloads or data to move between environments. Integration and governance are more complex.
• Community cloud: Infrastructure is shared by organisations with common requirements, such as regulatory or mission needs. It is less common than public, private and hybrid models.

Cloud adoption should be based on business requirements rather than fashion. Relevant criteria include workload variability, data sensitivity, resilience, latency, compliance, skills, integration, exit options and total cost of ownership.

6. Virtualisation

Virtualisation abstracts computing resources from physical hardware. A virtual machine (VM) is a software-defined computer containing virtual processors, memory, storage and network interfaces. A hypervisor creates and manages VMs. A Type 1 hypervisor runs directly on hardware; a Type 2 hypervisor runs on a conventional operating system.

Virtualisation allows several VMs to share one physical server while remaining logically isolated. It improves hardware utilisation, enables rapid provisioning, supports snapshots and simplifies testing and disaster recovery. However, VMs still consume physical resources, and excessive consolidation can produce contention. A failure in a host or management platform may affect many VMs, making resilience and monitoring essential.

Containers provide a lighter form of operating-system-level isolation. Unlike VMs, containers commonly share the host operating-system kernel. They start quickly and are useful for packaging applications and their dependencies. Container orchestration platforms automate deployment, scaling and recovery, but they introduce operational complexity and require careful image and privilege management.

Virtualisation is a foundational technology for cloud computing, but cloud computing additionally involves self-service, pooled resources, automation, elasticity and service measurement.

7. Edge Computing

Edge computing places processing, storage or analytical capability close to the location where data is produced or consumed, rather than sending all data to a distant central cloud. An edge location may be a factory gateway, retail branch, telecommunications site, vehicle or on-premises server.

Edge computing is useful when latency, bandwidth, privacy or intermittent connectivity matters. For example, a factory may analyse machine vibration locally and stop equipment immediately when a dangerous pattern is detected. Only summaries or selected events need to be sent to the cloud for long-term analysis.

Edge does not replace cloud computing. A common architecture is:

device or sensor → edge gateway → regional service or cloud → enterprise application

The edge performs rapid response and local filtering; the cloud supports large-scale storage, model training, coordination and historical analysis. Challenges include securing many distributed devices, maintaining software remotely, handling inconsistent data and ensuring that local decisions remain accurate.

8. Managerial and Technology-Sourcing Considerations

Technology sourcing is the process of deciding whether capabilities should be developed internally, purchased, leased, outsourced or delivered through a cloud service. A sound decision considers more than price.

Evaluation Criteria

1. Strategic importance: Capabilities central to competitive advantage may deserve internal control.
2. Cost: Compare acquisition, migration, licensing, staffing, support, energy, downtime and exit costs. Total cost of ownership is more informative than the initial price.
3. Performance: Consider latency, capacity, availability and scalability.
4. Security and privacy: Assess identity controls, encryption, logging, incident response and data location.
5. Compliance: Determine whether industry, contractual or national rules restrict processing or transfer.
6. Integration: Examine APIs, data formats, identity systems and legacy dependencies.
7. Vendor risk: Evaluate financial stability, subcontractors, concentration, service-level history and lock-in.
8. Reversibility: Establish data-export formats, transition assistance and termination rights.
9. Skills and operating model: Ensure the organisation can administer, monitor and govern the chosen service.
10. Sustainability: Consider energy consumption, hardware life cycles and provider environmental reporting.

A service-level agreement (SLA) should define measurable service commitments, maintenance windows, support response times, incident reporting, backup responsibilities, recovery objectives and remedies. Availability should be expressed carefully. If a service is promised at 99.9% monthly availability, the maximum unavailable time in a 30-day month is approximately:

This calculation assumes a 30-day month and that the percentage applies to the entire month. It does not by itself describe performance, security or data durability.

Business continuity planning should distinguish recovery time objective (RTO)—how quickly a service must be restored—from recovery point objective (RPO)—the maximum acceptable amount of data loss measured in time. A system with an RPO of 15 minutes requires a recovery design that limits potential data loss to approximately the previous 15 minutes; it does not necessarily mean backups occur exactly every 15 minutes.

Governance and Ethical Issues

Networked and cloud systems raise important governance questions. Organisations should collect only necessary personal data, define legitimate purposes, restrict access and retain information for appropriate periods. Encryption protects data in transit and at rest, but encryption does not correct poor identity management or excessive permissions.

Managers should apply least privilege, multi-factor authentication, network segmentation, secure configuration baselines, patch management, vulnerability testing, logging and tested backups. Contracts should identify breach notification duties, audit rights, subcontracting arrangements, data residency and deletion procedures.

Ethically, firms should consider digital exclusion, surveillance, algorithmic bias in cloud-hosted analytics, employee monitoring and the environmental effects of data-centre consumption. Automated systems deployed at the edge may make decisions affecting safety or customers; accountability must remain with identifiable organisational decision-makers.

Applied Case: Meridian Retail’s Connected Stores

Background

Meridian Retail operates 80 stores. Each store uses point-of-sale terminals, inventory scanners, security cameras and Wi-Fi for staff and customers. Its existing WAN is expensive, inventory data is synchronised overnight, and managers cannot easily identify equipment failures.

Problem

The company wants real-time stock visibility, reliable payment processing and predictive maintenance for refrigeration equipment. It must protect payment data, maintain service during temporary Internet interruptions and control costs.

Analysis

A purely centralised cloud design would simplify analysis but could create unacceptable dependence on continuous connectivity. Sending all camera and sensor data to the cloud would increase bandwidth and privacy risks. A private data centre would provide control but require substantial capital and specialist staff.

Approach and Solution

Meridian adopts a hybrid architecture. Each store receives a secure edge gateway that filters sensor data and continues essential local operations during short outages. Transaction and inventory applications run in a public-cloud PaaS environment, while sensitive payment functions use a specialised compliant service. Staff and customer WLANs are separated through network segmentation. Site-to-site encrypted connections link stores to cloud services, and multi-factor authentication protects administration.

The company negotiates an SLA covering availability, incident notification, data export and termination assistance. It tests restoration quarterly and measures latency, failed transactions, stock-synchronisation delay and recovery performance.

Outcome

The design reduces unnecessary traffic, improves response to refrigeration faults and provides more frequent inventory updates. However, Meridian must continue managing edge-device patches, cloud permissions, vendor relationships and integration complexity.

Learning Points

The case demonstrates that cloud adoption is an architectural and managerial decision, not simply a purchase decision. Hybrid cloud and edge computing can address latency and resilience, but they increase governance obligations. Network segmentation and shared-responsibility analysis are essential to protecting business services.

Challenges and Contemporary Developments

Important challenges include cyberattacks, ransomware, supply-chain compromise, misconfiguration, identity theft, outages, vendor lock-in and rapid technology obsolescence. Organisations increasingly use zero-trust principles, in which access is continuously verified rather than automatically trusted because a user is inside the corporate network.

Artificial intelligence is influencing network operations through anomaly detection, capacity forecasting and automated incident triage. It can improve speed but may generate false positives, conceal reasoning or amplify poor training data. Human oversight, audit trails and clear escalation procedures remain necessary.

Software-defined networking separates network control from individual hardware devices and enables central policy-based management. Network function virtualisation implements functions such as firewalls or load balancers in software. These approaches can improve agility, but they increase dependence on controllers, automation and skilled configuration.

Best practice is therefore balanced: design for resilience, secure identities and configurations, monitor actual performance, test recovery, document dependencies, retain exit options and align architecture with business value.

Chapter Summary

Computer networks connect devices and enable communication, resource sharing and coordination. LANs, WANs, wireless networks and the Internet differ in scale, ownership and use. Protocols such as IP, TCP, UDP, DNS, HTTP, HTTPS, DHCP and Ethernet provide the rules needed for interoperable communication. The Internet is a distributed network of autonomous systems, while the Web is one service that operates over it.

Wired networks offer predictable performance and physical control; wireless systems provide mobility but require careful radio planning and security. Cloud computing supplies pooled, elastic and measured resources through IaaS, PaaS and SaaS service models and public, private, hybrid and community deployment models. Virtualisation abstracts computing resources and supports efficient cloud infrastructure. Edge computing moves selected processing closer to devices to reduce latency, bandwidth use and dependence on continuous connectivity.

Managers must evaluate sourcing decisions using strategic value, total cost, performance, security, compliance, integration, vendor risk, reversibility and organisational capability. Effective governance includes shared-responsibility analysis, SLAs, identity protection, segmentation, monitoring, tested recovery and ethical attention to privacy, inclusion and accountability.

Key Terms

Network: Connected devices and communication links that exchange data.

Protocol: Rules governing communication between systems.

LAN: Local area network covering a limited location.

WAN: Wide area network connecting geographically separated networks.

Router: Device that forwards packets between networks.

Switch: Device that forwards frames within a local network.

IP: Protocol providing logical addressing and packet routing.

TCP: Transport protocol providing reliable, ordered delivery.

DNS: Distributed system that maps domain names to IP addresses.

Internet: Global interconnection of independent networks.

Cloud computing: On-demand delivery of pooled computing resources over a network.

IaaS: Cloud delivery of virtualised infrastructure resources.

PaaS: Cloud platform for developing and running applications.

SaaS: Cloud delivery of a complete software application.

Virtualisation: Abstraction of computing resources from physical hardware.

Hypervisor: Software that creates and manages virtual machines.

Edge computing: Processing located near data sources or users.

SLA: Agreement defining measurable service commitments.

RTO: Target time for restoring a service.

RPO: Maximum acceptable period of data loss measured in time.

Review Questions

Short Answer Questions

1. What is the difference between bandwidth, throughput and latency?
2. Distinguish a LAN from a WAN.
3. What functions do DNS and IP perform?
4. Define IaaS, PaaS and SaaS.
5. What is the purpose of an SLA?

Long Answer Questions

1. Explain the layered approach to networking and discuss why protocols are necessary.
2. Compare wired and wireless networking in terms of performance, mobility, security and management.
3. Explain public, private, hybrid and community cloud deployment models.
4. Describe virtualisation and explain its relationship to cloud computing.
5. Discuss the principal criteria managers should use when sourcing technology services.

Analytical / Application Questions

1. A business has real-time safety sensors in a factory with unreliable Internet connectivity. Recommend an architecture using edge and cloud computing and justify your design.
2. A firm is considering SaaS for its customer relationship management system. Analyse the security, integration, vendor-lock-in, cost and exit issues it should investigate.
3. A provider promises 99.5% monthly availability. Calculate the maximum unavailable time in a 30-day month and explain why this figure alone is insufficient for evaluating the service.

Multiple Choice Questions with Answers

Question 1

Which device primarily forwards packets between different networks?

A. Switch

B. Router

C. Access point

D. Printer

Correct Answer: B. Router

Question 2

Which protocol translates a domain name into an IP address?

A. DNS

B. DHCP

C. SMTP

D. Ethernet

Correct Answer: A. DNS

Question 3

Which cloud service model gives the customer the greatest responsibility for operating systems and applications?

A. SaaS

B. PaaS

C. IaaS

D. Community cloud

Correct Answer: C. IaaS

Question 4

Which is a characteristic of cloud computing?

A. Mandatory ownership of physical servers

B. On-demand self-service

C. Permanent fixed capacity

D. No resource measurement

Correct Answer: B. On-demand self-service

Question 5

What is the principal purpose of a hypervisor?

A. Translate domain names

B. Manage virtual machines

C. Encrypt email messages

D. Route Internet traffic between countries

Correct Answer: B. Manage virtual machines

Question 6

Which situation most strongly supports edge computing?

A. A historical archive with no time-sensitive processing

B. A factory requiring immediate local response to sensor data

C. A printed employee directory

D. A one-time software installation

Correct Answer: B. A factory requiring immediate local response to sensor data

Question 7

In the shared-responsibility model, which activity normally remains a customer responsibility?

A. Physical security of the provider’s data centre

B. Manufacture of the provider’s servers

C. Correct configuration of customer identities and permissions

D. Maintenance of the provider’s power supply

Correct Answer: C. Correct configuration of customer identities and permissions

Question 8

What is the main purpose of network segmentation?

A. To increase monitor brightness

B. To isolate systems and limit the spread of compromise

C. To replace all authentication

D. To eliminate the need for protocols

Correct Answer: B. To isolate systems and limit the spread of compromise

Question 9

A 30-day month has a service availability commitment of 99.9%. Approximately how much unavailable time does this permit?

A. 4.32 minutes

B. 43.2 minutes

C. 432 minutes

D. 4,320 minutes

Correct Answer: B. 43.2 minutes

Question 10

What does RPO measure?

A. The time required to authenticate a user

B. The maximum acceptable amount of data loss measured in time

C. The number of routers in a network

D. The monthly cost of a cloud service

Correct Answer: B. The maximum acceptable amount of data loss measured in time