Unit 5: Advanced IT Technologies

I. Orientation: The Shift to Distributed, Connected Computing

Advanced IT technologies move computing away from single, locally owned machines toward resources that are pooled, delivered over networks, and embedded in physical objects. Two ideas anchor this unit: cloud computing (commercial roots in the mid-2000s, e.g. Amazon Web Services launched 2006) delivers computing as a metered utility, and the Internet of Things (term coined by Kevin Ashton, 1999) extends the internet to everyday devices.

• On-demand delivery: Resources (compute, storage, network) are provisioned instantly and released when done, rather than bought as fixed hardware.
• Abstraction: The user works with a logical service and never sees the underlying physical machine.
• Measured service: Usage is metered (CPU-hours, GB-months, requests) and billed pay-as-you-go.
• Connectivity: Devices and services communicate over IP networks, generating and consuming data continuously.
• Scalability and elasticity: Capacity grows or shrinks with demand automatically.

II. Cloud Computing: Computing as a Utility

The delivery of computing services (servers, storage, databases, software) over the internet, so users rent capacity instead of owning infrastructure.

A. Introduction to Cloud Computing

The core purpose is to convert capital expenditure on hardware into operational expenditure on a service.

• Essential characteristics (NIST model): on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service.
• Deployment models:
  • Public cloud: Shared infrastructure owned by a provider (AWS, Azure, Google Cloud); lowest cost, least control.
  • Private cloud: Dedicated to one organisation, on-premises or hosted; more control and compliance.
  • Hybrid cloud: Combines public and private, moving workloads between them (e.g. sensitive data private, burst traffic public).
• Benefits: no upfront hardware, global reach, automatic updates, and resilience through provider redundancy.
• Concerns: dependence on internet connectivity, vendor lock-in, and data-sovereignty/security questions.

B. Cloud Service Models

Cloud offerings are layered by how much of the stack the provider manages versus the customer.

1. IaaS (Infrastructure as a Service): The provider supplies virtualised compute, storage, and networking; the customer manages the OS and everything above.
   • Example: Amazon EC2 virtual machines, where you choose CPU, RAM, and install your own OS.
   • Control vs burden: Maximum flexibility, but you patch the OS and configure the stack.
2. PaaS (Platform as a Service): The provider manages the OS, runtime, and middleware; the customer deploys only application code and data.
   • Example: Google App Engine or Heroku, where you push code and the platform runs it.
   • Trade-off: Faster development, less control over the environment.

• SaaS (Software as a Service): Fully managed application delivered through a browser; the customer manages only their data and settings (e.g. Gmail, Salesforce, Office 365).
• The stack analogy: Compare to eating — IaaS is buying ingredients and cooking, PaaS is a meal kit, SaaS is dining at a restaurant.

C. Cloud and Database Storage

The cloud stores both unstructured files and structured records, exposed as services that scale independently of compute.

• Object storage: Data kept as objects (file + metadata + ID) in a flat namespace, accessed by API; used for backups, media, and archives.
  • Example: Amazon S3 buckets, priced per GB-month with virtually unlimited capacity.
• Block and file storage: Block storage (e.g. EBS) attaches to a VM like a raw disk; file storage exposes a shared filesystem.
• Cloud database storage: Managed databases where the provider handles patching, backups, and replication.
  • Relational (SQL): Structured tables with fixed schema, ACID transactions — e.g. Amazon RDS running MySQL/PostgreSQL.
  • Non-relational (NoSQL): Flexible schema for large or varied data — key-value, document, or column stores such as DynamoDB or MongoDB Atlas.
• Advantages: automatic replication across data centres, durability (S3 advertises 99.999999999% durability), and elastic capacity.

III. Virtualization and Virtual Servers: The Engine Beneath the Cloud

Virtualization is the technology that lets one physical machine present itself as many independent logical machines, and it is what makes resource pooling in the cloud economical.

A. Concept of Virtualization and Virtual Servers

Virtualization creates a software abstraction of physical hardware so that several isolated environments share the same underlying resources.

• The hypervisor: A software layer that allocates CPU, memory, and I/O among virtual machines and keeps them isolated.
  • Type 1 (bare-metal): Runs directly on hardware — e.g. VMware ESXi, Microsoft Hyper-V; used in data centres for performance.
  • Type 2 (hosted): Runs on top of a host OS — e.g. VirtualBox, VMware Workstation; used on personal machines.
• Virtual server (VM): A software-defined computer with its own virtual CPU, memory, disk, and OS, running as a guest on shared physical hardware.
  • Isolation: A crash or compromise in one VM does not affect its neighbours.
• Containers (lightweight virtualization): Share the host OS kernel and package only the application and its dependencies — e.g. Docker; faster and smaller than full VMs.
• Types of virtualization: server, storage (pooling disks into one logical resource), and network virtualization.
• Benefits: higher hardware utilisation, faster provisioning, easy snapshots and migration, and lower cost and energy use.

Worked idea — consolidation:

1 physical server: 32 vCPU, 128 GB RAM
Hosts 8 virtual servers, each 4 vCPU / 16 GB RAM
Result: one box does the work of eight, raising utilisation

IV. Internet of Things (IoT): Connecting the Physical World

IoT is a network of physical objects embedded with electronics, software, and sensors that collect data and exchange it over the internet, often acting without human intervention.

A. Basics of IoT and Its Applications

The defining goal is to bridge the physical and digital worlds so that everyday objects become sources of data and points of control.

• "Thing": Any object given a unique identity and network connection — a thermostat, a wearable, a machine on a factory floor.
• Core capabilities: sense the environment, transmit data, process it, and act on the result.
• Communication protocols: MQTT (lightweight publish/subscribe messaging), CoAP, HTTP, plus link technologies like Wi-Fi, Bluetooth, Zigbee, and LoRaWAN chosen by range and power.
• Applications:
  • Smart home: connected lights, thermostats (Nest), security cameras controlled by app or voice.
  • Healthcare: wearables tracking heart rate; remote patient monitoring.
  • Industrial IoT (IIoT): predictive maintenance from vibration and temperature sensors on machinery.
  • Smart cities: connected traffic signals, smart parking, environmental monitoring.
  • Agriculture: soil-moisture sensors triggering irrigation.

B. IoT Framework

An IoT framework is the layered architecture that carries data from a physical device to a useful decision. A common four-layer model:

1. Perception (device) layer: Sensors and actuators interact with the physical environment — the point where data originates and action occurs.
2. Network (transport) layer: Moves data from devices to processing systems using gateways and communication protocols (Wi-Fi, cellular, MQTT).

• Processing (middleware) layer: Stores and analyses data, frequently in the cloud; applies rules, filtering, and machine learning to turn raw readings into insight.
• Application layer: Delivers results to the end user through dashboards, mobile apps, or automated controls (e.g. a smart-home app).
• Edge vs cloud processing: Edge computing analyses data near the device to cut latency and bandwidth; the cloud handles heavier storage and analytics.
• Cross-cutting concern — security: Each layer needs authentication and encryption, since exposed devices widen the attack surface.

C. Role of Sensors and Actuators in IoT Framework

Sensors and actuators sit in the perception layer and form the two complementary halves of physical interaction: one reads the world, the other changes it.

1. Sensors (input): Detect a physical quantity and convert it into an electrical/digital signal for the system to read.
   • Examples and quantities: temperature (thermistor, °C), humidity (DHT11, %), motion (PIR), light (LDR, lux), proximity, and accelerometers.
   • Function: They are the source of data — without sensing there is nothing to transmit or analyse.
2. Actuators (output): Receive a command and convert an electrical signal into physical action, closing the loop.
   • Examples: motors (rotate), relays (switch power on/off), servos (precise angle), solenoid valves (open/close flow), buzzers, and LEDs.
   • Function: They execute the decision the system reaches.

• The control loop: sensor reads → data sent through network layer → processing layer applies a rule → command returned → actuator acts.
  • Example: A temperature sensor reads 30°C, the rule "if > 28°C turn on fan" fires, and a relay actuator switches the fan on.
• Transducer: The umbrella term for both, since each converts energy from one form to another.
• Signal handling: Analog sensor outputs are converted by an ADC (analog-to-digital converter) before a microcontroller can process them; actuator commands often pass through a driver to supply adequate power.