Comprehensive Study Guide to Computer Systems, Hardware Architecture, and Operating Systems
Module Overview: CPS4001 Computer Systems and Networks
Curriculum Context:
- Module Code: CPS4001
- Module Title: Computer Systems and Networks
- Academic Institution: St Mary’s University, Twickenham, London
- Instructor: Dr Ranyin Wang
- Core Aim: Designed to equip students with a comprehensive understanding of foundational concepts and practical implementations of computer systems and network infrastructures.
Structure and Contact Hours:
- Duration: -week academic module.
- Weekly Contact Format: -hour lecture combined with a -hour laboratory practical each week.
Assessment Schedule:
- Assessment 1 (\ total weight): Computer-Based Test scheduled in Week 6, assessing all content taught across Weeks 1 through 5.
- Assessment 2 (\ total weight): Network Design & Packet Tracer Project, with the final deliverable due in Week 12.
Key Learning Objectives:
- Identify core physical and architectural components of a computer system and detail the precise function of each.
- Describe the mechanisms governing hardware and software interaction.
- Recognise the structural impact of Artificial Intelligence (AI) on modern computer system design.
Fundamentals and Architecture of Computer Systems
Definition of a Computer System:
- A computer system is an integrated combination of hardware, software, data, and people working collaboratively to collect, process, store, and communicate information.
- A complete computer system is not restricted to physical machinery alone; it intrinsically depends on executing software instructions, stored data, and active human users.
The Four Essential Operational Stages:
- The computational pipeline functions through four sequential stages:

* **Input**: Hardware interfaces acquire raw data and signals into the system (e.g., keyboard, microphone, sensor).
* **Process**: The Central Processing Unit (CPU) receives and executes operational program instructions.
* **Output**: The processed results are communicated and displayed to external systems or users (e.g., monitor screen, audio speaker, printer).
* **Storage**: Data is permanently preserved across non-volatile media (e.g., Solid-State Drive [SSD], Hard Disk Drive [HDD]).
- Taxonomy of Computer Systems:
- Desktop:
- Example: Office PC.
- Key Characteristics: High-performance output, stationary and fixed workspace installation.
- Laptop:
- Examples: MacBook, ThinkPad.
- Key Characteristics: Portable form factor, balanced trade-off between power efficiency, battery longevity, and processing capability.
- Server:
- Examples: Web server, database server.
- Key Characteristics: High-reliability infrastructure, always powered on ("always on"), engineered for concurrent multi-user load handling.
- Embedded System:
- Examples: Smart thermostat, Automated Teller Machine (ATM).
- Key Characteristics: Single-purpose dedicated computing, optimized for ultra-low power consumption.
- Mobile Device:
- Examples: Smartphone, tablet.
- Key Characteristics: Mobile computing optimized for battery conservation, employing touch-based Input/Output (I/O).
- Supercomputer:
- Example: Weather forecasting clusters.
- Key Characteristics: Highly scaled parallel processing architectures running complex distributed mathematical models.
- Desktop:
Hardware Architecture and the CPU
Classification of Physical Hardware Subsystems:
- Processing Units: CPU, Graphics Processing Unit (GPU).
- Primary Memory: Random Access Memory (RAM), CPU Cache.
- Secondary Storage: Solid-State Drive (SSD), Hard Disk Drive (HDD).
- Input Devices: Keyboard, mouse, camera, microphone.
- Output Devices: Monitor, speakers, printer.
- Interconnectivity: All distinct physical hardware sub-units connect, communicate, and distribute power via the motherboard.
The Central Processing Unit (CPU):
- Functioning as the operational brain of the computer system, the CPU is responsible for fetching, decoding, and executing program instructions.
- Core Hardware Specifications:
- Clock Speed: Measures the volume of operational execution cycles completed per second, expressed in gigahertz (). A clock speed of equates to (2 billion cycles per second).
- Cores: Independent processing cores engineered onto a single physical silicon chip; an increase in core count scales multitasking performance and parallel computation.
- Cache: Ultra-fast, low-latency static memory fabricated directly adjacent to or inside the CPU cores.
- Industry Examples: Intel Core i-series, AMD Ryzen, Apple M-series.
Multi-Core Microarchitecture and Cache Hierarchy:
- Modern multi-core central processing units distribute high-speed cache memory hierarchically across discrete processing cores:
- Level 1 Cache (L1): Fastest cache memory, dedicated on a per-core basis.
- Level 2 Cache (L2): Secondary ultra-low latency cache, dedicated on a per-core basis.
- Level 3 Cache (L3): High-capacity cache memory pool shared across all operational cores on the physical die.
- Access speed decreases, and capacity scales up, following the path: .
- Modern multi-core central processing units distribute high-speed cache memory hierarchically across discrete processing cores:
The Instruction Execution Cycle (Fetch-Decode-Execute-Store):
- The operational mechanism by which a processor executes instructions involves a four-stage loop:

* **Fetch**: The CPU retrieves the next pending program instruction from primary system memory (RAM).
* **Decode**: The Control Unit interprets the fetched instruction to determine the necessary operations and data pathways.
* **Execute**: The Arithmetic Logic Unit (ALU) performs the designated arithmetic, logic, or control operations.
* **Store**: The computed outcome of the operation is written back into an internal CPU register or system RAM.
* **Execution Speed**: The cycle iterates billions of times every single second, governed by the system clock frequency.
Memory Hierarchy, Primary RAM, and Storage Technologies
Memory and Storage Comparative Hierarchy:
- CPU Cache:
- Speed: Fastest tier in the computing hierarchy.
- Volatility: Volatile (data immediately dissipates when electrical power is disconnected).
- Typical Size: Quantified in megabytes ().
- Random Access Memory (RAM):
- Speed: Very fast data transfer rates.
- Volatility: Volatile (all stored states and active runtime buffers are lost upon power loss).
- Typical Size: Typically ranges between and .
- Functional Scope: Holds executing operating system processes, active programs, and working data sets currently in use.
- Solid-State Drive (SSD):
- Speed: Fast non-volatile secondary storage.
- Volatility: Non-volatile (retains data states indefinitely without electrical current).
- Typical Size: Typically ranges from to .
- Mechanism: Employs non-mechanical silicon flash memory chips containing no moving components.
- Hard Disk Drive (HDD):
- Speed: Slow secondary storage tier relative to solid-state alternatives.
- Volatility: Non-volatile.
- Typical Size: Ranges from to .
- Mechanism: Relies on electromechanical moving parts consisting of rotating magnetic platters and actuator arms; delivers low financial cost per gigabyte ().
- CPU Cache:
Memory Data Flow Pipeline:
- To execute program data stored permanently on secondary storage, data travels sequentially up the storage hierarchy:

* Movement path: .
* **Real-World Benchmark**: Utilizing an SSD cuts Windows operating system boot times from approximately on a traditional mechanical HDD down to approximately .
Motherboard Architecture, I/O Subsystems, and System Buses
- The Motherboard Architecture:
- The motherboard is the primary Printed Circuit Board (PCB) establishing the physical and electrical interconnections for all computing components:

* **CPU Socket**: Structural socket mount housing the physical processor.
* **Chipset**: Silicon management controller directing communication pipelines between the processor, system memory, and input/output peripherals.
* **RAM Slots (DIMM)**: Dual In-line Memory Module connection slots housing system RAM.
* **BIOS/UEFI Chip**: Dedicated non-volatile firmware chip tasked with hardware self-tests and initial system startup verification.
* **PCIe Slots**: Peripheral Component Interconnect Express high-bandwidth expansion buses for discrete components (e.g., GPU, NVMe SSDs, network interface cards).
* **External Ports**: Standardized communication input/output receptacles (e.g., USB, HDMI, Ethernet).
Input and Output Devices:
- Input Categorization (bringing external data into the system):
- Manual Input: Keyboard, mouse, touchscreen interfaces.
- Automatic Input: Barcode scanners, microphones, cameras, physical sensors.
- Biometric Input: Fingerprint readers, facial recognition camera hardware.
- Output Categorization (delivering processed outcomes to users):
- Visual Output: Display panels (Liquid Crystal Display [LCD], Organic Light Emitting Diode [OLED]).
- Audio Output: Acoustic devices such as speakers and headphones.
- Print Output: Inkjet printers, laser printers, additive 3D manufacturing printers.
- Data Output: Network transmissions routed through network interfaces.
- OS Driver Dependency: All physical input and output devices interface directly with the CPU through specialized software device drivers supervised by the Operating System.
- Input Categorization (bringing external data into the system):
The System Bus Architecture:
- A system bus is an electrical communication highway designed to transfer data, memory references, and operational control signals between internal hardware components:

* **Data Bus**: Carries the actual binary data payload between the CPU, RAM, and I/O devices.
* **Address Bus**: Carries the physical memory address signaling where specific data must be retrieved or written.
* **Control Bus**: Transmits synchronization, read/write commands, and control status flags across devices.
* **Bus Width**: Dictates the volume of bits transferred concurrently across parallel lanes (e.g., a -bit bus transmits of data in parallel).
* **Universal Serial Bus (USB)**: Standardized external serial bus interface used for connecting and communicating with external peripherals.
Software Systems and the Software Stack
Software Categorization:
- Software is defined as a structured collection of machine-readable instructions that instructs physical hardware on how to execute operations.
- Without software, hardware is completely inert—a CPU deprived of software instructions executes nothing.
- System Software:
- Role: Directly interfaces with and manages hardware, providing a functional foundation for running user-level applications.
- Examples: Operating systems, hardware device drivers, firmware.
- Application Software:
- Role: Performs direct functional tasks on behalf of the end user.
- Examples: Microsoft Word, Google Chrome, Adobe Photoshop.
Detailed System Software Classification:
- Operating System (OS): Comprehensive platform software (Windows, macOS, Linux) that manages physical hardware resources and provides a graphical or command-line interface.
- Device Driver: Hardware-specific software (printer drivers, GPU display drivers) that translates standardized OS directives into machine instructions specific to a peripheral device.
- Firmware: Low-level machine code (BIOS, UEFI) permanently embedded within hardware non-volatile flash chips to handle low-level device configuration.
- Utility Software: Maintenance and security tools (antivirus utilities, disk defragmenters/cleaners) that monitor and sustain operating system health.
The Multi-Layer Software Stack:
- Computer architectures structure interaction into hierarchical layers:
- User: The person interacting with the computer system.
- Application Layer: Software applications executed by the user (e.g., Word, Chrome, games).
- Operating System Layer: Resource broker and management system (e.g., Windows, Linux, macOS).
- Firmware / BIOS Layer: Hardware interface firmware and low-level drivers (e.g., UEFI, device drivers).
- Hardware Layer: Physical computational components (e.g., CPU, RAM, Secondary Storage).
- Principle of Abstraction: Every successive layer abstracts away the complexity of the layers beneath it; for instance, a user does not need to understand low-level binary code to execute web browsing tasks.
- Direct Access Isolation: Applications cannot interface with or modify hardware directly; all hardware resource requests must be mediated via services provided by the operating system.
- Computer architectures structure interaction into hierarchical layers:
Operating System Core Functions and Key Concepts
The Five Critical Functions of the Operating System:
- Process Management: Schedules execution time on the CPU, governing which programs execute and arbitrating access among multiple tasks.
- Memory Management: Allocates and reclaims physical RAM address blocks among concurrent programs, isolating processes to prevent application memory corruption.
- File System Management: Structures, indexes, navigates, and stores file directories across secondary storage media.
- Device Management: Facilitates low-level communication with internal and peripheral hardware via standardized device drivers.
- User Interface Provision: Provides accessible interaction environments via Command-Line Interfaces (CLI) or Graphical User Interfaces (GUI).
Key Operating System Terminology:
- Kernel: The central core component of the operating system that resides permanently in RAM and maintains privileged, low-level control over all hardware access.
- Scheduler: The specific internal OS subsystem algorithm that determines which executing thread or process receives access to the CPU next.
- Virtual Memory: A memory management implementation where secondary storage space (SSD/HDD) is dynamically configured to emulate physical RAM when active system memory is fully saturated.
- Deadlock: A concurrency failure condition in which two or more processes are permanently frozen because each is holding a resource that the other process requires to proceed.
The Complete Computer Boot Sequence
- Procedural Stages from Power-On to User Authentication:
- Stage 1 (Power On): Electrical power is initiated via the power supply unit, energizing the motherboard and components.
- Stage 2 (POST - Power-On Self-Test): The system firmware (BIOS/UEFI) executes automated hardware diagnostic checks to ensure operational integrity of the CPU, memory, and baseline hardware.
- Stage 3 (Storage Detection): The BIOS/UEFI searches configured boot priorities to find a bootable secondary storage medium.
- Stage 4 (Bootloader Initialization): The BIOS/UEFI executes the storage drive's bootloader program (e.g., GRUB, Windows Boot Manager), which loads the operating system kernel from disk into physical RAM.
- Stage 5 (OS Kernel Initialization): The operating system kernel assumes hardware control, loads fundamental device drivers, and mounts system file partitions.
- Stage 6 (System Services Start): Core background services launch, initiating network stacks, security services, background daemons, and graphical interface subsystems.
- Stage 7 (Login Screen): The graphical login display initializes, presenting authentication controls and rendering the computer ready for human interaction.
Artificial Intelligence in Modern Computer System Design
AI System Integration:
- Artificial Intelligence in computing refers to software and hardware architectures engineered to perform tasks historically requiring human cognitive intelligence.
- Key Drivers of Mainstream Integration:
- Exponential scaling in available training and inference datasets.
- Microarchitectural performance improvements in modern processors.
- Standardization and hardware integration of dedicated AI processing engines (such as Neural Processing Units [NPUs] and Tensor Cores).
Architectural Evolution: Edge vs. Cloud Computing:
- A critical architectural shift is actively underway: rather than relying exclusively on remote cloud data centers, AI inference workloads are increasingly executing locally on-device at the edge.
Execution Topologies of Common AI Applications:
- On-Device (Edge Architecture):
- Apple Face ID: Local biometric computer vision and neural authentication running entirely on-device for security and immediate response latency.
- Autocorrect / Predictive Text: Real-time natural language prediction models executing locally on mobile hardware without network dependency.
- Cloud Server (Centralized Architecture):
- Email Spam Filters: Machine learning classifiers running across remote mail servers with centralized access to global spam patterns.
- Bank Fraud Detection: High-throughput transactional inference models running on distributed banking cloud clusters evaluating live fraud risk.
- On-Device (Edge Architecture):