Computer System

Written and reviewed by Rohit Singh, Computer Science & Commerce faculty at UniDrill (MCA, Osmania University · 6 years teaching). Based on NCERT Informatics Practices Class XI, Chapter 1.

What happens when you press a key?

You tap the letter "A". A fraction of a second later it appears on the screen, and a copy of it sits somewhere that vanishes if the power cuts out. In that instant, at least five different parts of the machine took part: one converted your key press into binary, one held it temporarily, one decided what to do with it, one calculated, and one displayed the result. This chapter names those parts, explains why a computer needs several kinds of memory instead of one big one, and then looks at the software that tells the hardware what to do.

The computer system: four parts working as one

A computer is an electronic device that can be programmed to accept data (input), process it and give a result (output). A computer together with its additional hardware and software is a computer system. It can be anything from a large server to a laptop, tablet or smartphone.

The physical parts, called hardware, are the CPU, the primary memory, the input devices and the output devices (with secondary storage kept alongside). Software is the other half, and the chapter ends with it.

The flow in one line: input device → primary memory ↔ CPU (CU + ALU + registers) → output device, with secondary storage attached to memory for permanent keeping.

Inside the CPU: ALU, CU and registers

The CPU is the electronic circuitry that carries out the actual processing, which is why it is called the brain of the computer. It is also called the processor, and a CPU built on a single chip is a microprocessor. Physically it sits on one or more chips called integrated circuits (IC), made of semiconductor material.

The CPU works in a cycle. A program tells it what to do, it fetches the instructions and data from memory, performs arithmetic and logical operations, and stores the result back in memory.

Why registers exist. Registers are limited in size and number, but they are inside the CPU, so reaching them costs almost no time. The CPU keeps what it needs this instant in registers, which is exactly why they sit at the top of the speed ladder later in this chapter.

Pause and predict: a student says "the ALU decides which instruction runs next". Which unit actually does that job?

Check your answer: the Control Unit. The ALU only computes (add, compare, AND, OR). Deciding the sequence, interpreting the instruction and directing the data flow is the CU's job.

Input and output devices

An input device sends control signals and data into the computer and converts them into digital form. An output device takes digital information from the computer and converts it into a form people can understand.

Special devices make computing accessible: a braille keyboard lets a visually impaired user enter data, and a braille display shows text output to them. Voice search is another input route. Data you type is first held in main memory (RAM), and only stored permanently if you save it to secondary memory.

Why it matters. A printer is the most common device for hardcopy (paper output). A document saved on a hard disk or pen drive is a softcopy; print it and it becomes a hardcopy.

Evolution of computers: each invention fixed a limit

Computing grew from simple calculators to powerful data processors in a short time. Read the timeline as a chain of problems and fixes.

A punched card is a piece of stiff paper that stores digital data as holes at fixed positions.

Von Neumann architecture. It has a CPU for arithmetic and logic, a memory that stores both data and programs, input and output devices, and communication channels between them. ENIAC is the first binary programmable computer based on this architecture.

Beyond the textbook. For the exam: ENIAC is the computer built on the Von Neumann stored program idea, and the timeline dates above are what get asked. In reality: the modern phone in your hand still follows the same layout of processor, memory and input/output, just shrunk onto chips. Only the scale changed, not the idea.

From chips to your pocket. In the 1970s, Large Scale Integration (LSI) put a complete CPU on one chip, the microprocessor. In the 1980s Very Large Scale Integration (VLSI) packed around 3 million components onto a chip, and Super Large Scale Integration (SLSI) pushed density higher still. IBM introduced its first personal computer for home users in 1981, and Apple introduced Macintosh in 1984. Graphical User Interface (GUI) operating systems replaced the command line interface of UNIX or DOS and made PCs popular, and the growth of the World Wide Web in the 1990s made computers part of everyday life. Laptops, smartphones and tablets followed, and the next wave is wearables such as smart watches and smart lenses, plus smart appliances in the Internet of Things (IoT).

Moore's Law. In 1965 Intel co-founder Gordon Moore predicted that the number of transistors on a chip would double every two years while costs would be halved. This is why a phone today outperforms a room-sized machine of the past.

Beyond the textbook. For the exam: Moore's Law says transistor count doubles every two years. In reality: it is an observation about the industry's trend, not a law of physics, and chip makers now find it harder to keep shrinking transistors, so the pace has slowed. Why it matters: it explains why newer devices are cheaper and faster, and why engineers now also add more processor cores instead of only shrinking parts.

Units of memory

A computer works in binary, so the basic unit of memory is the bit, a single 0 or 1. Bits are grouped into words. A 4-bit word is a nibble (for example 1001) and an 8-bit word, two nibbles, is a byte (for example 01000110).

Why 1024 and not 1000? Memory is built from binary switches, so sizes come in powers of 2, and $2^{10} = 1024$ is the power of 2 closest to 1000. Every step up the table multiplies by $2^{10}$.

Worked example: converting to bytes. Convert 2 MB, 3.7 GB and 1.2 TB into bytes.

The common wrong step is to multiply by 1000 each time, which gives 2000000 bytes for 2 MB. That is the decimal shortcut used by some storage sellers, not the binary 1024 the chapter uses.

Beyond the textbook. For the exam: 1 KB = 1024 bytes, always. In reality: disk makers often label 1 GB as one billion bytes, so a drive sold as 500 GB shows up as a smaller number in the operating system. A separate set of names (kibibyte, mebibyte and so on) exists to mean exactly 1024 steps. Why it matters: if a question says "as per the binary system", the factor is 1024.

Types of memory

A computer needs memory to hold data and instructions while it processes them. When people say "memory" they usually mean the main or primary memory. Secondary memory (the storage device) holds data, instructions and results permanently for future use.

Why not one big memory? Fast memory is expensive and loses data when power goes. Cheap, permanent memory is slow. So computers use layers: a little very fast memory near the CPU and a lot of slow permanent memory far away.

RAM (Random Access Memory) is the main memory. It is volatile: it keeps data only while power is supplied. Whenever the computer starts or you open an application, the required program and data are loaded into RAM. It is faster than secondary storage.

ROM (Read Only Memory) is non-volatile: contents stay when power is off. It is a small, faster permanent store for contents that rarely change, such as the boot loader, the startup program that loads the operating system into RAM.

Cache memory is very high speed memory placed between the CPU and primary memory. RAM is faster than storage but not as fast as the processor, so the CPU would have to wait. Cache keeps copies of data from frequently used RAM locations, reducing the average time to access data. The CPU looks in the cache first. If the data is there, it is read from the cache (a cache hit), otherwise RAM is accessed.

Secondary memory is non-volatile, larger than primary memory, slower and cheaper. The CPU cannot access it directly: contents must first be brought into main memory.

The speed ladder, fastest to slowest: registers → cache → RAM → secondary memory.

Pause and predict: you buy a computer with RAM but no secondary storage. Can you install a software on it?

Check your answer: no. Installing means storing the software permanently, and RAM loses everything when power goes off. Without a secondary storage device there is nowhere for the installed files to live, so every start would begin empty.

Beyond the textbook. For the exam: there is one cache between the CPU and primary memory. In reality: modern processors have several cache levels (commonly called L1, L2 and L3), with the smallest and fastest one closest to the processing core. Why it matters: the idea is unchanged. A small fast store near the processor holds copies of whatever is used most.

HDD vs SSD. The Hard Disk Drive (HDD) was the standard storage device. The Solid-State Drive (SSD) supports a much higher data transfer speed. Pen drives and flash drives make transfer between computers simple.

Data capturing, storage and retrieval

Before a computer can process data, it must be captured, then stored in a file or database, and retrieved whenever it is needed again.

Data deletion and recovery

Data can be lost because a device crashes, a user deletes it by mistake, or malware or a hacker deletes it on purpose. Deleting data bit by bit is slow, so when you delete a file the system simply marks its address entry as free and shows the space as empty. The actual data is still there until something is written over it.

That leads to two facts. Recovery is possible only if the space has not been overwritten. And data recovery means retrieving deleted, corrupted or lost data from secondary storage.

Why it matters. Think of deleting as tearing out a library's index card but leaving the book on the shelf. Anyone who knows where to look can still read it. Only overwriting, or shredding, removes the book.

Beyond the textbook. For the exam: a deleted file can be recovered until its space is overwritten. In reality: on many modern SSDs a command called TRIM lets the drive clear freed blocks in the background, so recovery of deleted files is often less likely than on an old hard disk. Why it matters: never rely on "delete" for confidential data, whatever the drive. Use a data shredding tool before disposal.

Software

Software is a set of instructions and data that makes hardware functional. You cannot touch or view it physically. Hardware is of no use alone, and software cannot work without hardware. Software also acts as an interface between the user and the hardware, because we cannot instruct hardware directly.

Software has three broad categories:

System software, in three parts.

Why device drivers exist. New devices appear every day, each with its own behaviour. The operating system cannot know them all, so the job of controlling one device is handed to that device's driver, which acts like a translator between the OS and the device.

Application software has two kinds:

Pause and predict: your computer has no word processing software installed. Can you still use the computer? What if it has no operating system?

Check your answer: yes in the first case, no in the second. A computer can work without application software, but it cannot work without system software.

Proprietary, freeware and FOSS

Software is also classified by the terms on which its maker releases it.

A software can be freeware, open source or proprietary depending on the terms and conditions of the person or group that developed and released it.

Worked example: classify the software. For each item, say whether it is system software, programming tool or application software: (a) Compiler, (b) Assembler, (c) Ubuntu, (d) text editor used for code.

Exam traps

60-second revision

Where this chapter connects

• Next: Emerging Trends builds on this chapter's hardware vocabulary to explain AI, cloud and the Internet of Things.
• Same course: Understanding Data picks up the capture, storage and retrieval story and asks what to do with the data once you have it.
• Python link: Brief Overview of Python uses a language that is itself FOSS, and its interpreter is a programming tool of the kind this chapter classified.