Computer Fundamentals & Architecture (Core Theory)

Computer Fundamentals & Architecture

Complete study notes covering computer fundamentals, hardware architecture, I/O devices, computing generations, Von Neumann architecture, CPU functional units, memory hierarchy, system buses, and worked numerical examples.

Contents

1. Computer fundamentals and data processing
2. Types of computers
3. Hardware and input/output devices
4. Characteristics of computers
5. Applications of computers
6. Advantages and disadvantages
7. Evolution and generations
8. Von Neumann architecture
9. Functional units and component interaction
10. Memory and storage
11. Computer buses
12. Worked numerical examples
13. Essential revision and answer practice
14. Essential examination traps and core distinctions
15. Linux laboratory practicals and command reference
16. Unit 1 modular study guides and examination modules

1. Computer fundamentals and data processing

1.1 What is a computer?

A computer is an electronic device that accepts data as input, processes it according to instructions, stores data and results, and produces output.

The word computer comes from compute, meaning to calculate. Arithmetic calculation is one use of a computer, but computers also store, retrieve, compare and otherwise process data.

A computer is therefore also called a data processor. Its work follows instructions supplied through software programs. The Central Processing Unit (CPU) carries out the processing.

Answer to learn: A computer is an electronic device that accepts data, processes it according to a program, stores information and produces results as output.

1.2 Five basic functions

Function	Meaning	Example in a marks-processing task
Input	Accept data and instructions	Enter a student's marks
Storage	Hold data, instructions and results for use when needed	Keep the entered marks and the calculation instructions
Processing	Perform operations that turn data into useful information	Calculate the total and average
Output	Present the processing result	Display the total and average
Control	Coordinate the other functions and their order	Arrange input, calculation, storage and display in the required sequence

Control is the fifth fundamental function. A complete architectural classification specifies input, processing, output, storage, and centralized control coordination.

1.3 Data, information and data processing

• Data: raw facts supplied as input.
• Information: data that has been processed into a useful result.
• Data processing: the activity of processing data using a computer.

The fundamental data-processing sequence is: capture data, manipulate data, output results.

Point	Data	Information
Form in the processing task	Raw input	Processed result
Role	Material to work on	Meaningful output of the work
Example	Marks: 70, 80 and 90	Total: 240; average: 80

Worked illustration: The values 70, 80 and 90 are the input data. Adding them gives 240. Dividing 240 by 3 gives an average of 80. The reported total and average are information produced by processing those data.

The same computer can carry out a different task when it receives a different program. The processing hardware and the instructions that direct it work together.

2. Types of computers

Computers are broadly classified in two ways: size and computing power, and method of operation (operational principle). Keep these classifications separate in an answer.

2.1 Classification by size and power

Type	Core Characteristics	Practical Use Cases & Examples
Supercomputer	Extremely powerful systems for very large computational problems	Scientific calculations, climate research and weather forecasting
Mainframe	Large systems for very large transaction workloads and many users; the lecture also emphasizes security	Banks and large companies
Minicomputer	Mid-range systems between mainframes and microcomputers, supporting multiple users	Business tasks
Microcomputer	Small, relatively affordable computers for individual use	Desktops, laptops, tablets and smartphones

Supercomputer versus mainframe: the lecture emphasizes intensive scientific computation for supercomputers and large transaction workloads for mainframes. These are different workload priorities.

Minicomputer versus microcomputer: the lecture's minicomputer category emphasizes a mid-range, multi-user system. Its microcomputer category emphasizes personal use.

Mainframes are enterprise servers engineered to handle very large, concurrent transaction workloads with high reliability and fault tolerance.

2.2 Classification by operation

Type	Operational Nature	Primary Applications & Use Cases
Analog computer	Works with continuously varying quantities, such as voltage or pressure	Physical simulations and measuring instruments
Digital computer	Works with discrete values, commonly represented in binary as 0 and 1	Laptops, phones and servers
Hybrid computer	Combines analog and digital capabilities	Medical and industrial systems

A hybrid computer combines the real-time speed of analog systems with the high-precision memory and storage of digital computers. The central idea to learn is the combination of continuous-signal processing and digital capabilities in one system.

Comparison example: “Laptop” describes a type of personal computer. “Digital” describes how it represents and processes information. Both descriptions can apply to the same device.

2.3 Key Hardware Subsystems and Chassis Components

The image identifies embedded computer, laptop, server, mainframe computer, supercomputer and workstation.

These labels must be recognized, but the image does not give a detailed definition of every category. In particular, it does not establish a separate specification or performance threshold for servers and workstations. The embedded-system role is discussed again under the architecture applications in Section 8.8.

3. Hardware and input/output devices

3.1 Components shown in the hardware image

Key system components include CPU, motherboard, RAM, cooling system, power supply unit, and secondary storage.

Label	What to recognize
CPU	The processor, explained in Sections 8 and 9
Motherboard	The primary printed circuit board housing CPU, memory, and expansion bus connections
RAM	Fast volatile primary system memory directly accessible by the CPU
Cooling system	Heat sinks and fans dissipating thermal energy generated by processor operations
Power supply unit	Regulates and converts AC mains electricity into low-voltage DC power for internal circuitry
Hard disk	High-capacity non-volatile magnetic or solid-state secondary data storage

Technical Distinction: In computer hardware engineering, the CPU is the microprocessor integrated circuit installed on the motherboard, whereas the external housing is the system cabinet/unit.

3.2 Input devices

An input device allows data or instructions to enter the computer system.

The standard categories of computer input devices include:

Device	Primary Input Function
Keyboard	Alphanumeric text, numerical data, and command entry
Mouse	Point-and-click graphical cursor control and navigation
Touchscreen	Direct fingertip/stylus tactile gesture interaction
Scanner	Optical image and printed document digitization
Microphone	Audio signal and acoustic wave capture for digitization
Webcam	Real-time video streaming and photographic capture
Joystick	Multi-directional 2D/3D navigation and simulation control
Trackball	Stationary pointing device with rotating surface sphere
Light pen	Light-sensitive wand for drawing and screen coordinate selection
Optical card reader	Magnetic/smart card data decoding
Optical barcode reader	Universal Product Code (UPC) and barcode data scanner
Optical mark reader (OMR)	High-speed optical mark and examination sheet sensor

The diagram says optical card reader, not optical character recognition. Do not silently replace that label with a different term or assume both have the same full form.

3.3 Output devices

An output device presents the computer's processed results to the user or the outside world.

Valid output devices include:

• Monitor or Visual Display Unit (VDU), including LCD monitors.
• Printer.
• Speakers.
• Headphones and earphones.
• Projector.
• Plotter.
• The display part of a touchscreen.

A sound card acts as an internal audio controller that translates digital signals into analog sound output through connected speakers or headphones.

Note: An optical mark reader (OMR) is strictly an input device that senses marked data on answer sheets, not an output peripheral.

3.4 Why a touchscreen appears in both lists

A touchscreen combines an output display with an input surface that accepts touch. It therefore serves dual input and output roles.

3.5 Input versus output

Point	Input	Output
Direction relative to the computer	Into the system	Out of the system
Purpose	Supply data or instructions	Present results
Example	Entering data with a keyboard	Viewing results on a monitor

The full functions of the input and output units appear in Section 9.

4. Characteristics of computers

Computer architecture defines seven foundational characteristics:

4.1 Automatic operation

After a task has been set up with the necessary program and data, a computer can perform its programmed steps automatically without a person directing every step.

Example: once marks and calculation instructions are supplied, the computer can calculate totals and display results automatically.

4.2 Speed

Computers perform computational cycles at extremely high speeds, measured in standardized sub-second time units:

Unit	Symbol	Value in seconds
Microsecond	μs	$10^{-6}$ s, or one millionth of a second
Nanosecond	ns	$10^{-9}$ s, or one billionth of a second
Picosecond	ps	$10^{-12}$ s, or one trillionth of a second

Thus 1 microsecond = 1,000 nanoseconds, and 1 nanosecond = 1,000 picoseconds. These conversions follow directly from powers of ten ($10^{-6}$, $10^{-9}$, $10^{-12}$).

These are units of time. They are not a claim that every computer operation takes the same time.

4.3 Accuracy

Computers perform calculations with high precision; the level of operational accuracy is determined by hardware architecture and arithmetic logic unit design.

Correct operation also needs suitable instructions and input: a computer processing an incorrectly entered mark can faithfully calculate a result from that incorrect value. Automatic processing does not make every supplied value correct.

4.4 Versatility

A computer can perform many different kinds of tasks when each task can be expressed as a finite sequence of logical steps.

The same general-purpose computer can support calculation, document preparation, communication and other applications through different programs.

4.5 Power of remembering

Computers can store information and retrieve it later, especially through secondary storage.

A foundational characteristic of modern computing is large storage capacity and the ability to recall stored information reliably from internal and external media.

4.6 “No I.Q.”

Computers lack autonomous judgment; they execute exactly the instructions provided by software algorithms. While programs perform conditional logic and intelligent processing, operations depend fundamentally on deterministic programmed procedures.

4.7 “No Feelings”

Computers operate free of emotional biases, stress, or fatigue. Operations and outputs follow strictly verified algorithmic instructions.

4.8 Quick recall

Automatic operation, speed, accuracy, versatility, remembering, no I.Q., no feelings.

An explanation should accompany each heading when the question asks you to describe the characteristics.

5. Applications of computers

5.1 Communication

• Email and messaging: Gmail, Outlook, WhatsApp and Slack.
• Video conferencing: Zoom, Google Meet and Microsoft Teams.
• Social media: Facebook, Instagram, X (Twitter) and LinkedIn.

These applications allow people to exchange messages, participate in meetings and share information.

5.2 Education and learning

• Online classes and e-learning: Coursera, Udemy and Swayam-NPTEL.
• Research and study: Google Scholar, Wikipedia and digital libraries.
• Interactive learning: simulations, virtual labs and educational games.

The central role is access to learning material, research information and interactive study activities.

5.3 Business and work

• Office productivity: word processors, spreadsheets and presentations.
• Data management: databases and Customer Relationship Management (CRM) systems such as Salesforce and HubSpot.
• Automation: payroll systems, inventory control and AI chatbots.

Computers help organizations prepare documents, maintain records and automate routine work.

5.4 Banking and finance

• Online banking: checking account balances, transferring money and paying bills.
• Digital payments: PayPal, Google Pay and Apple Pay.
• Stock trading: Robinhood and E-TRADE, as named in the lecture.

These are examples of processing financial information and transactions.

5.5 Healthcare

• Medical records: Electronic Health Records (EHR).
• Diagnostics and imaging: MRI, CT scans and X-rays.
• Telemedicine: online doctor consultations and health-monitoring devices.

5.6 Entertainment

• Streaming: Netflix, YouTube and Spotify.
• Gaming: PC games and virtual-reality experiences.
• Digital art and music production: Photoshop and FL Studio.

5.7 Shopping

• E-commerce: Amazon, eBay and local online marketplaces.
• Price comparison: tools for comparing available deals.
• Order tracking and delivery services.

5.8 Transport and navigation

• Ride-hailing: Uber and Lyft.
• GPS navigation: Google Maps.
• Traffic information and public-transport updates.

5.9 Home automation

• Smart devices: Alexa, Google Home and smart thermostats.
• Security: CCTV monitoring and smart locks.
• Appliance control: smart ovens and washing machines.

5.10 Creative work

• Design and media production: video editing, 3D modelling and animation.
• Writing and publishing: blogs, books and newsletters.
• Photography and image editing.

5.11 Engineering and industrial design

• Computer-Aided Design (CAD): 2D drafting and 3D geometric modeling of mechanical parts, civil structures, and architecture (AutoCAD, SolidWorks, CATIA).
• Computer-Aided Manufacturing (CAM): Computer numerical control (CNC) machining, robotic assembly lines, and 3D printing toolpaths.
• Simulation and Finite Element Analysis (FEA): Stress analysis of bridges, aerodynamics testing using Computational Fluid Dynamics (CFD), and vehicle crash simulations.
• Electronic Design Automation (EDA): Designing, simulating, and routing printed circuit boards (PCBs) and microchip VLSI layouts (SPICE, KiCad).

How to write an applications answer: name the field, explain the task that computers perform, and add one or two relevant examples. A list of brand names alone does not explain the application.

6. Advantages and disadvantages

6.1 Advantages

Advantage	Explanation
Speed and efficiency	Large amounts of data can be processed quickly, with automation reducing repetitive work and saving time
Accuracy	Correctly configured computations and tasks can produce precise results and reduce manual calculation errors
Storage capacity	Large amounts of information can be organized and stored in a relatively compact physical space
Multitasking	A system can support multiple ongoing tasks, improving productivity
Versatility	The same system can support applications in education, business, entertainment and other fields

Multitasking clarification: the prescribed course describes both sharing processor time and using multiple CPUs. Several ongoing tasks need not execute on one CPU at exactly the same instant. CHO source: NPTEL course 106105214, Lecture 3.

6.2 Disadvantages

Disadvantage	Explanation given in the lecture
Health issues	Prolonged use can contribute to eye strain, back pain and repetitive-strain problems
Cost	Setup, maintenance and software updates can be expensive
Dependency	Excessive reliance may reduce practice of manual skills and independent critical thinking
Security risks	Inadequately secured systems can be exposed to hacking, viruses and data breaches
Technical issues	Hardware or software failures can interrupt work and may require specialist help

These are possible disadvantages and risks, not outcomes that must occur every time a computer is used.

7. Evolution and generations

7.1 Meaning of evolution

Evolution of computers means their overall development over time. It includes changes in devices, hardware, software, capability and usage.

It is the continuous story of how computing developed, rather than a single technological stage.

7.2 Key Historical Computing Milestones

Foundational pioneers and milestone mechanical computing devices include:

Pioneer / Device	Era	Landmark Computing Contribution
Blaise Pascal (Pascaline)	1642	First mechanical adding machine using geared toothed wheels
Baron Gottfried Wilhelm von Leibniz	1671	Stepped Reckoner mechanical calculator performing direct multiplication
Charles Babbage (Difference Engine)	1822	Designed automatic polynomial calculation engine
Charles Babbage (Analytical Engine)	1842	First general-purpose programmable design featuring an ALU (mill) and memory (store)
Keyboard Keypunch Machines	~1880	Mechanical alphanumeric data entry units developed in the United States
Herman Hollerith	~1880	Punched-card tabulating system used for the 1890 US Census, founding precursor to IBM
Punched Cards as Input Media	Through late 1970s	Standard batch-processing storage and program loading medium

Charles Babbage is universally honored as the father of computing because his Analytical Engine conceived the separation of storage from arithmetic processing, anticipating the stored-program computer.

7.3 Meaning of a computer generation

A computer generation is a stage of development identified by a major technological change.

Computer generations classify hardware architectures and system software paradigms across five major evolutionary eras:

7.4 Five-Generation Comparison

Generation	Era	Core Circuit Technology	Programming Paradigm	Landmark Computer Models
First	1940s–1956	Vacuum tubes	Machine language, represented in binary	ENIAC, UNIVAC I
Second	1956–1963	Transistors	Assembly language	IBM 1401, CDC 1604
Third	1964–1971	Integrated circuits (ICs)	High-level languages, including COBOL and FORTRAN	IBM 360 series, PDP-8
Fourth	1970s–present	Microprocessors	High-level languages such as C and C++	Apple II, IBM PC
Fifth	Present and future	AI, machine learning and parallel processing	Python, Prolog and natural-language processing as an approach	Siri, ChatGPT and supercomputers

Curriculum Note: Generational transitions reflect dominant hardware circuit technologies (Vacuum Tubes → Transistors → ICs → Microprocessors → Parallel/AI Cores).

7.5 First generation

• Used vacuum tubes.
• Programs used machine language.
• Systems were very large, often occupying a room.
• Produced substantial heat and consumed considerable electricity.
• First-generation machines operated at millisecond speeds, comparatively slow relative to modern microprocessors.
• Examples: ENIAC and UNIVAC I.

Full forms:

• ENIAC: Electronic Numerical Integrator and Computer.
• UNIVAC I: Universal Automatic Computer I.

7.6 Second generation

• Transistors replaced vacuum tubes.
• Assembly language introduced symbolic mnemonics (e.g., ADD, MOV) replacing binary machine code.
• Systems became smaller than first-generation machines.
• They were faster and more reliable, produced less heat and consumed less power.
• Examples: IBM 1401 and CDC 1604.

Key comparison: the transition from vacuum tubes to transistors improved size, reliability, speed and power requirements.

7.7 Third generation

• Used integrated circuits, combining many transistors on a chip.
• High-level languages (COBOL, FORTRAN) enabled platform-portable scientific and business computing.
• Systems became cheaper, smaller and more efficient.
• Business and government use expanded in the lecture's account.
• Examples listed: IBM 360 series and PDP-8, subject to the model-classification caveat above.

The appearance of a language in this row does not mean that the language first came into existence during that date range.

7.8 Fourth generation

• Used microprocessors, bringing a CPU onto a single chip.
• High-level structured languages (C, C++) accompanied microprocessor-driven personal workstations.
• Personal computers became common.
• Networking grew rapidly.
• Examples: Apple II and IBM PC.

Technical correction: a CPU on one chip is not the same statement as every component or function of an entire computer being on that chip. Memory, storage and input/output remain distinct roles in the system diagrams.

7.9 Fifth generation

• Focuses on artificial intelligence, machine learning and parallel processing.
• Emphasizes high speed and large-scale data processing.
• Includes systems concerned with learning and computational decision-making.
• AI-oriented programming (Python, Prolog) and natural language processing drive intelligent agents and neural networks.
• Examples listed: Siri, ChatGPT and supercomputers.

This is the lecture's AI-oriented generation category. It overlaps with computers that continue to use microprocessors.

7.10 Evolution versus generation

Aspect	Evolution	Generation
Meaning	Overall development from early computing tools to modern systems	A specific stage defined by a major technological change
Scope	Broad: hardware, software, applications and usage	Narrower: the defining technology of a stage
Focus	Continuous progress and trends	Distinct phases and breakthroughs
Time span	The whole history of computing	The period associated with one generation
Example	Abacus, mechanical calculators, ENIAC and modern AI across a long history	Vacuum-tube, transistor and IC stages
Nature	Explains how and why computers developed	Classifies computers by the technology emphasized in each stage

Recall sequence: vacuum tubes, transistors, integrated circuits, microprocessors, AI-oriented systems.

8. Von Neumann architecture

8.1 Definition and name

Von Neumann architecture is a computer-organization model in which instructions and data are stored in the same main memory and the processor retrieves instructions to execute them.

Also known as the Princeton architecture, it was formalized by mathematician John von Neumann in 1945.

The key point is a stored program: the instructions describing the task are kept in memory, where the processor can obtain them.

8.2 Historical attribution in the lecture

ENIAC creators John Mauchly and J. Presper Eckert, working alongside John von Neumann, developed the stored-program concept in which instructions and program data share the same physical memory space.

While named after John von Neumann, the stored-program paradigm was a collaborative architectural milestone involving Mauchly, Eckert, and the EDVAC engineering team.

8.3 Fixed-program versus stored-program computers

Point	Fixed-program computer	Stored-program computer
Task	Designed for a specific function	Can carry out different tasks through different programs
Changing the task	Limited or no user reprogramming in the lecture's example	A different stored program supplies different instructions
Instructions	Hardwired to a fixed set of mechanical/electrical steps	Stored in memory and easily reprogrammed
Example or role	Dedicated fixed hardware (e.g. four-function calculator)	General-purpose computing workstations

The qualification non-programmable matters. The lecture's calculator example should not be extended to every calculator ever made.

8.4 Key characteristics

1. Stored-program concept: memory contains the program's instructions.
2. Shared memory: both data and instructions occupy the same main-memory system.
3. Sequential instruction execution in the basic model: instructions are fetched and executed in sequence, with the next location determined by program control.
4. Control-unit direction: the CU coordinates execution and data movement.
5. Shared communication path: the basic model uses the same memory communication system for instruction and data transfers.
6. Common control mechanism: A centralized Control Unit orchestrates both data transfers and instruction decoding using synchronized system clock signals.

“Sequential” describes the basic execution model. It does not require every instruction to be at the next numeric address regardless of jumps.

8.5 CPU components and registers

A register is a storage location within the CPU used during processing. Different registers have different jobs.

Component or register	Full name	What it does
ALU	Arithmetic Logic Unit	Performs arithmetic and logical operations
CU	Control Unit	Coordinates CPU operations and the flow of instructions and data
PC	Program Counter	Holds the address of the next instruction to be fetched
IR	Instruction Register	Holds the current instruction being decoded or executed
MAR	Memory Address Register	Holds the address of the memory location involved in an access
MDR	Memory Data Register	Temporarily holds the contents being transferred to or from memory
Accumulator	Accumulator register	Holds intermediate results of arithmetic and logical operations
General-purpose registers	General-purpose registers	Temporarily hold working data during processing

The word data in MDR's name does not mean it can never carry an instruction. In the shared-memory model, instruction bits also arrive as contents transferred from memory.

Four distinctions to learn:

• PC versus IR: PC holds an instruction's address; IR holds the instruction.
• MAR versus MDR: MAR holds the memory location; MDR holds the transferred contents.
• CU versus ALU: CU directs operations; ALU performs arithmetic and logic.
• Accumulator versus main memory: the accumulator is a CPU register; main memory is a separate storage component in the system diagrams.

8.6 Fetch–decode–execute cycle

Stage	Explanation	Main components involved
Fetch the instruction	Use the PC to locate the next instruction and bring it from memory into the CPU	PC, memory, CU and the instruction-transfer path
Decode	Interpret the instruction to determine the requested operation	CU and IR
Fetch operands when needed	Obtain the data required by the instruction and place them in CPU registers	Memory, registers and the data-transfer path
Execute	Carry out the requested operation	ALU or another appropriate part of the system
Place the result	Put the result in a register or memory, as required by the instruction	Registers or memory
Continue	Update or use the PC for the next instruction and repeat	PC and CU

An operand is a value on which an operation acts. In addition, the two values being added are operands.

The instruction execution cycle operates across four fundamental stages: fetch, decode, execute, and writeback/repeat.

Decode does not mean translating a program into a new high-level language. It means interpreting the instruction and determining the actions or control signals needed to execute it.

PC Timing: In modern CPU pipelines, the Program Counter (PC) increments immediately during the fetch stage to point to the subsequent sequential memory address, unless a branch or jump instruction overrides it.

Worked illustration: adding two values

This example illustrates the architectural instruction cycle flow common to general-purpose processors.

1. Memory holds an instruction requesting the addition of 7 and 5.
2. The PC identifies where that instruction is stored.
3. The CPU fetches it, and the CU decodes the requested operation as addition.
4. The required values are available in registers, fetching them from memory if needed.
5. The ALU calculates 7 + 5 = 12.
6. The result is placed in the destination register or memory location specified by the operation.
7. Execution continues with the next instruction.

Instruction versus operand: “add” describes the operation. The values 7 and 5 are the operands. The value 12 is the result.

8.7 Advantages and limitations

Architectural Advantages

Advantage	Explanation
Simplified design	A unified memory arrangement reduces the complexity of having separate instruction and data memories in the basic model
Cost-effectiveness	The simpler arrangement can reduce component and implementation costs
Flexibility	Different stored programs allow different tasks on the same general-purpose system
Ease of programming	A common memory model provides a consistent organization for programs and their data
Wide adoption	The model provides a common conceptual foundation for computer organization and software education

Wide architectural adoption established standard computer design principles, though software cross-compatibility requires an identical Instruction Set Architecture (ISA).

Von Neumann bottleneck

The Von Neumann bottleneck is the limitation on transfer between CPU and main memory caused by their shared communication path.

Instructions and data compete for that path. In the simple single-path model, it cannot fetch both independently at the same time. The CPU may therefore spend time waiting for the next instruction or required data, limiting useful throughput.

Cause and effect: shared path, competing instruction/data transfers, waiting, reduced throughput.

Architectural Limitations

Limitation	Meaning in the basic model
Shared-bus limitation	Instruction and data transfers compete for the same path
Memory bottleneck	Limited transfer rate or memory response time can leave the CPU waiting
Sequential processing	The simple model handles instructions in a sequence and does not itself provide a mechanism for parallel execution
Scalability issues	High-performance work demanding frequent memory access can intensify the transfer limitation
Energy inefficiency	Frequent movement between CPU and memory contributes to energy use
Latency	Delays in obtaining instructions or data increase completion time

Modern Computing Note: While the classic Von Neumann model executed instructions strictly sequentially, modern processors achieve parallelism through instruction pipelining, superscalar execution, and multi-core architectures.

8.8 Practical Applications & Everyday Systems

• General-purpose computing: the conceptual foundation taught for desktops, laptops and smartphones.
• Embedded systems: simple devices where cost and design simplicity matter.
• Software development: an organization model that influences programming tools and languages.
• Education: a foundational model used to explain computer architecture.
• Gaming and multimedia: support for programs such as games and editing software.

These are the lecture's application areas. They do not claim that every real device implements the elementary diagram without additional features.

9. Functional units and component interaction

9.1 Main functional units

A digital computer operates on data through coordinated functional units:

1. Input unit.
2. Output unit.
3. Storage or memory unit.
4. Central Processing Unit, including the CU and ALU.

At a high functional level, the CPU integrates the Control Unit (CU) and Arithmetic Logic Unit (ALU) alongside internal register banks.

9.2 Block diagram

The following diagram illustrates the functional relationships between the CPU and system components. Solid connections show data movement, while dotted connections represent control relationships from the CPU's control unit.

flowchart TB
    I["Input unit"] --> P["Primary storage"]
    P --> O["Output unit"]
    P <--> S["Secondary storage"]
    P <--> C["CPU: CU, ALU and registers"]
    C -.-> I
    C -.-> O
    C -.-> P

Read the diagram as a functional model. It does not specify every wire, bus cycle or possible device-transfer method.

9.3 Input-unit functions

The input unit:

1. Accepts data and instructions from an input device.
2. Represents that input in a form the computer can handle.
3. Participates in transferring the input into the processing system.
4. Stores incoming data in main memory for CPU access.

Input units transfer external signals into memory buffers across the system bus, while the Control Unit maintains command over system execution.

Input hardware digitizes external physical signals into computer-readable binary format (binary data words), which are then stored for software processing.

Examples: keyboard, mouse, microphone and scanner. Section 3 retains every input label shown in the images.

9.4 Output-unit functions

The output unit:

1. Receives the results of processing.
2. Converts or presents them in a form supported by an output device.
3. Communicates the results to users or the outside world.

Output processing is the functional inverse of input conversion: internal binary states are decoded and transformed into human-readable visual displays, physical printouts, or acoustic signals.

9.5 Storage-unit functions

The storage unit holds:

• Data and instructions awaiting processing.
• Intermediate results produced during processing.
• Final results awaiting output.
• Data and programs retained for later use.

Intermediate result: a value obtained while a larger calculation is still in progress.
Final result: the completed result of that task.

The unit includes the primary- and secondary-storage categories described in Section 10.

9.6 Central Processing Unit

The CPU executes program instructions and coordinates computation and data flow, functioning as the primary computational engine of the computer.

Its ALU performs arithmetic and logic. Its CU directs the required actions. Its registers hold instructions, addresses, data and intermediate results during processing.

9.7 Control Unit

The Control Unit (CU) serves as the computer's central coordinator: it fetches, decodes, and routes timing signals to all operational units.

Its roles include:

• Receiving instruction-related information from memory.
• Interpreting what must happen next.
• Coordinating the ALU, memory, input and output.
• Controlling the order and flow of operations.

Architecturally, the CU does not perform mathematical transformations or long-term data storage; arithmetic calculations are performed by the ALU, while operands reside in registers and memory.

9.8 Arithmetic Logic Unit

The ALU performs:

• Arithmetic: addition, subtraction, multiplication and division.
• Logical comparisons: greater than, less than and equal to.

It carries out operations selected under CU control. It does not independently manage the overall instruction sequence.

Point	CU	ALU
Main role	Direct and coordinate	Calculate and compare
Works with	Instruction interpretation and control signals	Operands and operation results
Example	Arrange an addition and the required data transfers	Add 7 and 5 to obtain 12
Relationship	Specifies the required operation	Executes the requested arithmetic or logical operation

9.9 How the units cooperate

For a marks-processing program:

1. The input unit accepts the marks.
2. Memory holds them and the program instructions.
3. The CU directs the required transfers and operations.
4. Registers hold working values.
5. The ALU calculates the total or performs a comparison.
6. The result can be placed in memory and sent to the output unit.
7. Secondary storage can retain a saved result for later use.

This example connects the units rather than treating them as unrelated definitions.

10. Memory and storage

10.1 Primary memory

Primary memory provides high-speed working storage directly accessible by the CPU to hold currently executing machine instructions and active operands.

The lecture emphasizes that working memory is generally:

• Directly accessible by the CPU in the basic model.
• Faster than secondary storage.
• Smaller in capacity than secondary storage in the comparison being taught.
• Limited in capacity.

Primary memory encompasses both Random Access Memory (RAM) and Read-Only Memory (ROM).

RAM: Random Access Memory. The working RAM discussed here holds active data and programs and is volatile.
ROM: Read-Only Memory. It retains its contents without power and can hold startup instructions.

Key Distinction: RAM is volatile (temporary) and loses its contents upon power loss; ROM is non-volatile (permanent) and preserves system boot routines (BIOS/UEFI) across restarts.

10.2 Secondary memory

Secondary storage provides permanent, non-volatile data retention with larger capacity, lower cost per gigabyte, and persistent storage when electrical power is disconnected.

The word permanent in the lecture means storage that survives ordinary power-off. It does not mean the information can never be deleted or lost.

10.3 Comprehensive Memory & Storage Classification

Category shown	Labels in the diagram
Primary memory	RAM, ROM
Secondary memory	“Punching Device,” magnetic tape, floppy disk, optical disk, hard drive, flash disk

Punched Media: Early systems used punched paper cards and paper tape as physical recording media, read by mechanical or optical sensors.

10.4 Primary versus secondary storage

Point	Primary memory	Secondary storage
Role	Active instructions/data and other directly accessible memory in the teaching model	Longer-term retention of programs and files
CPU access	Direct in the basic model	Normally requires transfer into working memory for execution
Speed	Generally faster	Generally slower
Capacity in the lecture comparison	Usually smaller	Usually larger
Cost per unit of storage	Generally higher	Generally lower
Power-off behaviour	Depends on type: RAM is volatile; ROM is non-volatile	Non-volatile
Storage Examples	RAM (SRAM, DRAM) and ROM (PROM, EEPROM)	HDD, SSD, USB flash drives, Optical Discs, Magnetic Tape

Programs and files retained in secondary storage are brought into main memory when needed for execution. CHO support: IISc NPTEL Module 4.

10.5 CHO supplement: memory hierarchy

The standard computer memory hierarchy spans multiple performance tiers, balancing access latency, storage density, and cost per bit:

Level, from closest/fastest working storage downward	Main purpose
CPU registers	Hold very small amounts of immediate working information inside the CPU
Cache	Keep information that can serve repeated accesses more quickly than main memory
Main working memory	Hold active programs and data
Secondary storage, including hard disk	Retain larger collections of programs and files

Across this hierarchy, faster levels generally have smaller capacity and a higher cost per bit. Lower levels offer larger capacity at lower cost per bit, but take longer to access. A system combines them to balance those needs. Source: prescribed NPTEL course 106105214, Lecture 4.

Cache and repeated access

Cache is a relatively small, fast store. If required information is already there, the system can use that copy without repeatedly consulting a slower level. Which information is retained and how frequently it is reused affect the benefit.

Access time is the time required to obtain information. Bandwidth is the amount transferred per unit time. Low access time and high bandwidth both help, but they describe different properties. The numerical capacities shown in the course are examples, not fixed specifications for every computer. Source: prescribed NPTEL course 106105214, Lecture 6.

The course also illustrates a multicore arrangement with an L1 cache for each core and a shared L2 cache. This demonstrates multiple cache levels, without making that layout universal. Source: prescribed NPTEL course 106105214, Lecture 5.

10.6 Volatile versus non-volatile

Question	Volatile working storage	Non-volatile storage
Does it preserve its contents when power is removed?	No	Yes, under ordinary power-off conditions
Example relevant here	Working RAM	ROM and secondary storage
Why is it needed?	Supports work in progress	Preserves information for later use

Hierarchy and volatility answer different questions. Hierarchy compares storage levels and their performance/capacity trade-offs. Volatility concerns power-off retention. A primary-memory label by itself is not enough to determine volatility when the category includes ROM.

Reference: Principles of Computer Organization & Memory Hierarchy.

11. Computer buses

11.1 What is a bus?

A computer bus is a communication link connecting components such as the CPU, memory and input/output hardware.

The system bus comprises three distinct functional signal pathways:

• Data bus.
• Address bus.
• Control bus.

Together, they answer three questions: what is being transferred, which location is involved, and what operation should occur?

While motherboard PCB traces also deliver DC electrical power, architectural bus theory classifies the system bus into three logical signal groups: Data Bus, Address Bus, and Control Bus.

11.2 Data bus

The data bus transfers contents between the CPU, memory and other components.

It is bidirectional in the lecture's model:

• During a memory read, information moves from memory toward the CPU.
• During a memory write, information moves from the CPU toward memory.

Data-bus width is the number of bits that the bus can carry in one transfer in the simple parallel-bus examples.

Width	Data carried in one transfer
8 bits	8 bits
16 bits	16 bits
32 bits	32 bits
64 bits	64 bits

Data bus width determines single-cycle transfer throughput (e.g. 32-bit or 64-bit parallel word transfers between the CPU and memory controller).

11.3 Address bus

The address bus identifies the memory location involved in a read or write.

In the simple CPU-controlled model, it is unidirectional: the CPU sends the address to memory or another addressed component.

The width determines the number of distinct addresses, not the number of data bits carried in one transfer.

For an address bus of n bits:

For ordinary unsigned numbering beginning at zero:

Crucial assumption: an address identifies a location. To express capacity in bytes, the number of bytes stored at each addressed location must also be known.

If each address selects one byte, then:

More generally:

In standard microprocessor architectures, the address bus is unidirectional from the CPU to memory, though Direct Memory Access (DMA) controllers can also drive address lines during autonomous peripheral transfers.

11.4 Control bus

The control bus carries signals that coordinate the operation and timing of connected components.

The lecture mentions:

• Indicating a read or write operation.
• Synchronizing component actions.
• Carrying response signals from hardware.
• Communication through the motherboard and other physical connections.

The control-bus group is described as bidirectional because some control signals go from the CPU to devices and other signals return from devices. That does not mean every individual signal line must carry signals both ways.

Control Bus Signal Propagation: Individual control lines are unidirectional: command pulses originate from the CPU, while status/acknowledgement signals flow back from memory and I/O devices.

11.5 Three-bus comparison

Bus	Direction in the basic model	Carries	Main question answered	Key role
Data	Bidirectional	Actual transferred contents	What is being transferred?	Moves information between components
Address	Unidirectional from the CPU	Address of the location	Where is the information to be read or written?	Selects the location
Control	Signals in both directions across the group	Commands, responses and synchronization signals	What operation should occur, and when?	Coordinates CPU and device actions

11.6 Memory read

To read a memory location in the conceptual model:

1. The CPU identifies the required location, represented in the MAR.
2. That address is sent on the address bus.
3. A read signal is sent through the control bus.
4. Memory supplies the contents on the data bus.
5. The received contents are available through the MDR for the required CPU use.

Roles to remember: the address bus selects, the control bus requests, and the data bus carries the result.

11.7 Memory write

To write to a memory location:

1. The CPU identifies the destination address.
2. It supplies the contents to be stored.
3. The address is presented on the address bus.
4. The contents are presented on the data bus.
5. A write signal on the control bus directs memory to store those contents at the selected address.

These steps explain the memory transfer process. They provide a conceptual transfer description, not a claim about exact signal timing for a particular processor.

11.8 I/O interfaces in the diagrams

In standard system architecture, an I/O bus, I/O interfaces and connected devices are organized in series. The interface sits between the bus-side connection and a device, forming the complete communication path between peripheral hardware and the system bus.

11.9 Data width versus address width

Question	Relevant width
How many bits can this parallel data transfer carry at once?	Data-bus width
How many unique locations can be selected?	Address-bus width
How many bytes can be addressed?	Address width plus bytes per addressed location

A 32-bit data bus does not, by itself, prove that the address bus is also 32 bits. The two widths describe different jobs.

12. Worked numerical examples

12.1 Units used in the calculations

• 1 byte = 8 bits.
• 1 KiB = 1,024 bytes.
• 1 MiB = 1,024 KiB = 1,048,576 bytes.

Standard computer architecture calculations use binary multiples ($1\text{ KB} = 2^{10} = 1,024\text{ bytes}$, $1\text{ MB} = 2^{20} = 1,048,576\text{ bytes}$).

12.2 Four-bit address bus

Question: How many locations can a 4-bit address bus select?

The addresses range from 0 to 15. If each address selects one byte:

This yields a maximum block transfer capability of 16 bytes.

12.3 Sixteen-bit address bus

Question: What capacity can a 16-bit address bus address in a byte-addressable system?

Answer: 65,536 byte locations, or 64 KB ($2^{16}$). The valid address range is 0 to 65,535.

12.4 Twenty-bit address bus

With one byte per location:

Answer: 1 MB ($2^{20} = 1,048,576\text{ bytes}$). The valid address range is 0 to 1,048,575.

12.5 When a location contains more than one byte

Derived illustration: suppose a 4-bit address bus selects locations that each store 2 bytes.

It still selects 16 locations, but those locations collectively store 32 bytes. This shows why the byte-addressability assumption must be stated.

12.6 Finding address width from capacity

Derived illustration: how many address bits are required for 256 KiB of byte-addressable memory?

One byte per address means $2^{18}$ distinct addresses are required.

Answer: 18 address bits.

12.7 Data-bus width and transfer size

Data-bus width	Calculation	Bytes per full transfer
8 bits	8 ÷ 8	1 byte
16 bits	16 ÷ 8	2 bytes
32 bits	32 ÷ 8	4 bytes
64 bits	64 ÷ 8	8 bytes

Derived illustration: to transfer 128 bits over a 32-bit data bus using complete transfers:

This counts transfers. It does not determine elapsed time without additional timing information.

12.8 Common numerical mistakes

1. Writing n locations for an n-bit address bus instead of 2ⁿ locations.
2. Giving a byte capacity without stating the bytes per location.
3. Confusing bits with bytes.
4. Treating data-bus width as address-bus width.
5. Using 65,536 as the final address in a 65,536-location space that begins at zero. The final address is 65,535.
6. Misinterpreting exponential powers of two (e.g. reading $2^4 = 16$ as twenty-four).

13. Essential revision and answer practice

13.1 Compact revision sheet

Topic	Essential answer
Computer	Electronic device that accepts, processes and stores data, and produces output under program control
Five functions	Input, storage, processing, output and control
Data	Raw facts used as input
Information	Useful result of processing
Size/power categories	Supercomputer, mainframe, minicomputer, microcomputer
Operation categories	Analog, digital, hybrid
Seven characteristics	Automatic, speed, accuracy, versatility, remembering, no I.Q., no feelings
Evolution	Overall development over time
Generation	A stage defined by an important technological change
Generation technology sequence	Vacuum tubes, transistors, ICs, microprocessors, AI-oriented systems
Von Neumann name	Also called Princeton architecture
Stored-program principle	Instructions are stored in memory for execution
Shared memory	Instructions and data use the same memory in the model
Instruction cycle	Fetch, decode, obtain operands when required, execute, place result and continue
PC	Next instruction's address
IR	Current instruction
MAR	Address involved in a memory access
MDR	Contents transferred to or from memory
Accumulator	Intermediate arithmetic/logic results
CU	Coordinates operations
ALU	Calculates and compares
Von Neumann bottleneck	Instructions and data compete for the CPU–memory transfer path
Data bus	Transferred contents; bidirectional
Address bus	Selected location; CPU-originated direction in the simple model
Control bus	Commands, responses and synchronization
n address bits	2ⁿ unique addresses
Hierarchy to hard disk	Registers, cache, main memory, secondary storage

13.2 Short answers to understand

1. Why is a computer called a data processor?
Because it accepts data, performs operations on it, stores or retrieves it, and produces useful results.

2. Why can a computer be fast but still produce an incorrect result?
Speed describes how quickly operations occur. An incorrect input or instruction can still lead quickly to an incorrect result.

3. What is the central difference between a supercomputer and a mainframe in the lecture?
The supercomputer discussion emphasizes very large scientific calculations; the mainframe discussion emphasizes heavy transaction workloads.

4. How is versatility different from multitasking?
Versatility means suitability for different kinds of task. Multitasking means supporting multiple ongoing tasks.

5. What changed between the first and second generations?
Transistors replaced vacuum tubes, yielding major advances in compactness, operational reliability, computational speed, and reduced heat dissipation.

6. What is the key fourth-generation development?
The microprocessor, which places a CPU on a single chip.

7. Why does stored-program organization provide flexibility?
Different stored instruction sequences allow the processor to carry out different tasks.

8. Why is the PC not the same as the IR?
The PC identifies where to obtain the next instruction. The IR holds the instruction currently being handled.

9. What is the difference between an instruction and an operand?
An instruction specifies an operation. An operand is a value used by that operation.

10. Why are the CU and ALU both needed?
The CU coordinates the operation and the required transfers. The ALU performs the arithmetic or logical work.

11. Why does a fast CPU still wait for memory?
It may require an instruction or data that have not yet arrived. Limited transfer capacity and memory latency can delay processing.

12. Why is cache useful?
It can supply repeatedly needed information more quickly than consulting a slower storage level each time.

13. Why is a touchscreen both an input and an output device?
It accepts touch input and displays visual output.

14. Which buses take part in a memory read?
The address bus identifies the location, the control bus signals a read, and the data bus returns the contents.

15. Does an n-bit address bus necessarily address 2ⁿ bytes?
It selects 2ⁿ locations. The answer is 2ⁿ bytes only when each location holds one byte.

16. Does a wider data bus automatically increase addressable memory?
No. Data width describes the amount moved per transfer. Addressable locations depend on address width.

13.3 Longer answers: what to include

These are answer structures for revision, not a predicted exam pattern.

Question	Content to include
Explain the basic functions of a computer	Definition, all five functions and one connected example
Explain the characteristics of computers	All seven lecture headings, with explanations and appropriate qualifications
Classify computers	Separate size/power classification from operation-based classification; add examples
Discuss applications	Explain the fields and the actual tasks performed; include concrete industrial examples
Explain the five generations	Periods as taught, technology, languages, examples and changes between generations
Distinguish evolution from generation	Meaning, scope, focus, time span, examples and nature
Explain Von Neumann architecture	Stored program, shared memory, components, instruction cycle and bottleneck
Explain the CPU registers	PC, IR, MAR, MDR, accumulator and general-purpose registers, with their distinct roles
Draw and explain the computer block diagram	Input, output, primary/secondary storage, CPU, CU, ALU and the data/control relationships
Compare primary and secondary memory	Purpose, access, speed, capacity, cost, examples and volatility with the ROM qualification
Explain the three buses	Contents, direction, purpose and a memory read/write example
Solve an address-bus problem	Identify n, calculate 2ⁿ locations, state location size, convert units and give address range if asked

13.4 Abbreviations used in the notes

Abbreviation	Full form
CPU	Central Processing Unit
CU	Control Unit
ALU	Arithmetic Logic Unit
PC, in a CPU-register question	Program Counter
PC, in the computer-type context	Personal Computer
IR	Instruction Register
MAR	Memory Address Register
MDR	Memory Data Register
RAM	Random Access Memory
ROM	Read-Only Memory
I/O	Input/Output
VDU	Visual Display Unit
IC	Integrated Circuit
AI	Artificial Intelligence
EHR	Electronic Health Records
CRM	Customer Relationship Management
ENIAC	Electronic Numerical Integrator and Computer
UNIVAC I	Universal Automatic Computer I

Use the question's context to distinguish the two meanings of PC.

14. Essential Examination Traps & Core Distinctions

The following summary highlights key architectural distinctions frequently tested in university examinations:

Topic / Concept	Common Student Misconception	Core Architectural Principle
System Cabinet vs. CPU	Labeling the external computer chassis as the "CPU"	The CPU is the microprocessor chip installed on the motherboard; the case is the system unit/cabinet.
Memory Volatility	Assuming all primary memory is temporary	RAM is volatile (temporary) and clears on power-off; ROM is non-volatile (permanent) and stores boot firmware.
Programmable vs. Fixed	Assuming all automated machines are stored-program	Fixed machines (calculators, mechanical adders) lack software reprogrammability; stored-program architectures store instructions in RAM.
Control Bus Flow	Assuming control bus lines are bidirectional	Individual control lines are dedicated and unidirectional: command strobes flow out from the CPU, while status/interrupt signals flow inward.
Memory Capacity Calculation	Equating address locations directly with bits	By standard convention, primary memory is byte-addressable ($2^N$ addressable locations $= 2^N$ bytes).
Optical Mark Readers	Classifying an OMR as an output peripheral	An Optical Mark Reader (OMR) is strictly an input sensor that detects marked data on answer sheets.
Bus Masters	Assuming only the CPU can initiate memory addresses	In advanced systems with Direct Memory Access (DMA), the DMA controller drives addresses directly during high-speed I/O transfers.

15. Linux Laboratory Practicals & Command Reference

This appendix provides the core practical laboratory curriculum covering Linux shell commands, file systems, permissions, and GCC compilation.

15.1 Ubuntu installation outline

The prescribed tutorial uses a text-menu Ubuntu Server installer:

1. Prepare installation media and back up existing data that must be kept.
2. Start the computer from the installation USB drive or DVD.
3. Choose language, keyboard layout and installation option.
4. Configure networking.
5. Choose the storage device and inspect the proposed partitions.
6. Confirm changes only after checking the intended disk.
7. Set the user profile, hostname and password.
8. Complete installation, remove installation media and restart.

Disk confirmation can overwrite existing contents. The tutorial's screens describe its illustrated workflow; different Ubuntu releases may present the options differently. Prescribed Ubuntu tutorial.

15.2 GCC: compile and run

The CHO-linked video's accessible description provides these commands:

gcc --version
sudo apt install build-essential
gcc -o hello hello.c
./hello

Command	Purpose
gcc --version	Check the compiler version
sudo apt install build-essential	Install the build tools used in the tutorial
gcc -o hello hello.c	Compile hello.c into an executable named hello
./hello	Run the executable

Compilation produces the executable. Execution runs it. Prescribed GCC video description.

15.3 Timestamps and Makefiles

make checks a target's dependencies and their modification times. A source file that is newer than its dependent output can require that output to be rebuilt. This avoids unnecessary recompilation when files are already up to date.

A rule gives a target, its dependencies and the command needed to build it. Combining the prescribed GCC command with the NPTEL rule structure gives:

hello: hello.c
	gcc -o hello hello.c

The command starts with a tab. Save it in Makefile, then use:

make
./hello

Changing hello.c makes the dependent executable need rebuilding. The prescribed notes also explain:

• make -n: display commands without executing them.
• make -f filename: select a different makefile.
• touch filename: update a file's modification time.

Sources: IISc NPTEL Module 15 and the prescribed GCC video description.

15.4 File commands

Command or example	Purpose
ls	List directory contents
ls -l	Display a detailed listing
ls -R	Include subdirectories recursively
ls -t	Sort by modification time
ls -S	Sort by file size
cat notes.txt	Display file contents
cat -n notes.txt	Number the displayed lines
cp notes.txt copy.txt	Copy the file; an existing destination can be overwritten
cp -r source backup	Copy a directory recursively
mv old.txt new.txt	Move or rename a file
rm old.txt	Remove a file
mkdir -p revision/fcs	Create the directory path, including necessary parents

Prescribed Linux Documentation Project: Basic Commands.

15.5 Directory commands and paths

Command or symbol	Meaning
pwd	Display the absolute pathname of the working directory
cd notes	Change to the notes directory
cd ..	Change to the parent directory
cd	Return to the home directory
mkdir revision	Create a directory
rmdir revision	Remove an empty directory
.	Current directory
..	Parent directory
~	Home directory
/ inside a pathname	Separates pathname components

Sources: prescribed directory-management guide and, for the empty-directory requirement, the prescribed basic-commands reference.

Practical Scope: In addition to primary file and directory utilities, students should be familiar with standard paging utilities (less, more), search utilities (whereis, which), and basic shell customization.

16. Unit 1 Modular Study Guides & Examination Modules

To support systematic exam preparation, the Unit 1 curriculum is organized into modular study guides:

Module	Focus Area	Contents Included	Link
Module 01	Core Theory	Computer fundamentals, classifications, hardware, 5 generations, buses, worked numericals	Current Document
Module 02	Von Neumann Architecture	Deep dive into stored-program concepts, Harvard comparison, registers, bottleneck & 30 categorized MCQs	Open Topic Spotlight →
Module 03	MCQ Examination Bank	Complete 240 categorized questions covering Fundamentals, Generations, Architecture, CPU & Buses, Memory, Storage, Linux & GCC	Open MCQ Question Bank →
Module 04	Linux Practicals & Commands	5 hands-on lab programs with step-by-step terminal execution, expected output, stat breakdown & complete command manual	Open Linux Practicals →