Memory, Storage and Input–Output Technologies

Introduction

A computer system is useful only when it can temporarily hold instructions and data, preserve information for future use, accept input from users or other systems, and communicate results. These functions are performed by memory, storage, and input–output (I/O) technologies. Although the terms are often used interchangeably in everyday conversation, they describe different parts of a computer system with different purposes, speeds, costs, and reliability characteristics.

For example, a student may save a report on a solid-state drive (SSD), open it in a word processor, and edit it while the processor repeatedly accesses data held in random-access memory (RAM). Frequently used instructions may be copied into cache memory, while a portion of the SSD may be used as virtual memory when RAM is insufficient. The keyboard and mouse provide input, whereas the monitor and printer provide output. Understanding these relationships is essential for selecting, using, and managing business information systems.

Learning Objectives

After studying this chapter, students should be able to:

1. Distinguish primary memory, cache memory, virtual memory, and secondary storage.
2. Explain the hierarchy of computer memory and its effect on performance.
3. Classify input, output, and peripheral devices and describe their uses.
4. Compare magnetic, optical, solid-state, and cloud storage.
5. Evaluate storage alternatives using capacity, speed, reliability, portability, and cost.
6. Apply these criteria to practical organizational technology decisions.

1. The Memory and Storage Hierarchy

Computer memory can be understood as a hierarchy. At the highest levels, storage is very fast, expensive per byte, and located close to the processor. At lower levels, storage is slower but offers much greater capacity and lower cost per byte.

A simplified hierarchy is:

Processor registers → Cache → Main memory (RAM) → Secondary storage → Tertiary or cloud archives

The hierarchy exists because no single technology simultaneously provides maximum speed, capacity, permanence, and low cost. A computer therefore combines technologies and moves data between them according to current needs.

A memory device is volatile if its contents disappear when power is removed. RAM and cache are volatile. A device is non-volatile if it retains data without continuous electrical power. SSDs, hard disk drives, optical discs, magnetic tape, and many cloud storage services use non-volatile storage, although cloud availability also depends on network and provider access.

Data is commonly measured in bytes. A byte contains eight bits. In computing, capacity is often expressed in kilobytes, megabytes, gigabytes, and terabytes. Manufacturers commonly use decimal units, where 1 GB = 1,000,000,000 bytes. Operating systems may also report capacity using binary units, where 1 GiB = 1,073,741,824 bytes. The distinction explains why the capacity shown by an operating system may appear slightly smaller than the capacity printed on a storage package.

2. Primary Memory

2.1 Meaning and purpose

Primary memory, also called main memory or internal memory, is the memory directly accessible by the processor for currently running programs and active data. In modern general-purpose computers, primary memory is mainly RAM. When a user opens a spreadsheet, the operating system loads the relevant program and data from secondary storage into RAM. The processor then accesses the working copy in RAM much more quickly than it could access a disk.

Primary memory supports active processing rather than long-term preservation. Its capacity influences how many applications can run simultaneously and how much data can be processed without relying heavily on virtual memory.

2.2 RAM types

Dynamic RAM (DRAM) stores each bit using an electrical charge in a memory cell. Because the charge gradually leaks, DRAM must be refreshed repeatedly. DRAM is relatively dense and economical, so it is widely used as main memory.

Static RAM (SRAM) stores bits using electronic switching circuits and does not require the same periodic refresh process. It is faster but more expensive and less dense than DRAM. Consequently, SRAM is commonly used for cache rather than large main-memory modules.

RAM is normally random access: the system can access any address in approximately the same way, rather than reading all preceding data first. This differs from sequential-access media such as magnetic tape.

2.3 Read-only memory

Read-only memory (ROM) is non-volatile memory used to store instructions that should remain available when power is removed. Firmware—low-level software that controls or initializes hardware—is commonly stored in a form of ROM or flash memory. Modern firmware can often be updated, so “read-only” is a historical description rather than an absolute limitation.

3. Cache Memory

Cache memory is a small, high-speed memory located close to or within the processor. It stores copies of data and instructions that the processor is likely to use again soon. Cache reduces the time the processor spends waiting for slower RAM.

Cache works because of two important forms of locality:

• Temporal locality: recently used data or instructions are likely to be used again.
• Spatial locality: data or instructions located near a recently accessed address are likely to be used soon.

Processors generally use several cache levels. L1 cache is usually the smallest and fastest. L2 is larger but slower, and L3 may be larger still and shared among processor cores. Exact arrangements vary by processor design.

A cache hit occurs when requested information is already in cache. A cache miss occurs when it is absent and must be fetched from a lower memory level. A useful conceptual expression for average memory access time is:

Here, T_avg is average access time, h is the cache hit ratio, T_c is cache access time, and T_m is the effective time for a miss, including retrieval from the next level. This simplified expression assumes one cache level and treats miss handling as a single average time. In a multilevel cache, the calculation must include each level and its probability of being reached.

For example, suppose cache access takes 2 ns, the miss-handling time is 50 ns, and the hit ratio is 0.90. Then:

T_avg = 0.90(2) + 0.10(50) = 1.8 + 5 = 6.8 ns

The result illustrates why even a small cache can greatly improve performance. A high hit ratio is valuable because a cache miss may require access to slower memory.

Cache is not a replacement for large-capacity storage. It improves access speed by keeping selected information close to the processor; it does not permanently save user files.

4. Virtual Memory

Virtual memory is a memory-management technique that allows the operating system to use part of secondary storage as an extension of RAM. The operating system divides virtual memory into fixed-size blocks, commonly called pages, and transfers pages between RAM and a disk-based area called a page file or swap space.

Virtual memory provides each process with a logical address space that may be larger than the available physical RAM. It also improves protection by keeping processes in separate address spaces. A program can therefore behave as if it has a large, continuous memory area even when the physical memory is divided among many programs.

When a required page is not in RAM, a page fault occurs. The operating system must locate the page on storage, possibly move another page out of RAM, and load the required page. Because storage is much slower than RAM, frequent page faults can cause thrashing, in which the system spends more time moving pages than executing programs.

Virtual memory therefore provides capacity and isolation, but it is not an effective substitute for sufficient physical RAM. Adding RAM generally improves performance more reliably than increasing the size of the page file. Solid-state storage can reduce page-transfer delays compared with magnetic disks, but it remains slower than RAM and repeated writing may contribute to device wear.

5. Secondary Storage

Secondary storage retains programs and data for long-term use. It is normally non-volatile, has greater capacity than primary memory, and is less expensive per byte. Operating systems, applications, documents, databases, photographs, and backups are all commonly stored on secondary storage.

Secondary storage is often evaluated using the following criteria:

• Capacity: the amount of information that can be stored.
• Speed: how quickly data can be read or written; access time and throughput are important.
• Reliability: the likelihood of retaining data and operating without failure.
• Portability: how easily the device can be transported or used across locations.
• Cost: purchase price, maintenance, energy, replacement, and management costs.

5.1 Hard disk drives

A hard disk drive (HDD) stores data magnetically on rotating platters. A read/write head moves across the platter surface. HDDs offer high capacity at relatively low cost and remain useful for desktops, servers, and large collections of less frequently accessed data.

Their mechanical parts create disadvantages. Access time is affected by rotational delay and head movement. HDDs are also vulnerable to shock, vibration, and mechanical wear. They may consume more energy and produce more noise than solid-state alternatives.

5.2 Solid-state drives

A solid-state drive (SSD) stores data in flash memory and contains no rotating platters. SSDs generally provide much faster access, lower noise, lower power consumption, and better resistance to physical shock than HDDs. These characteristics make them valuable in laptops, business systems, and applications requiring rapid access.

Flash memory has a finite number of write and erase cycles. SSD controllers use techniques such as wear leveling, error correction, and spare blocks to extend useful life. SSD failure can also be sudden, so backups remain essential. An SSD’s speed may be limited by its interface, controller, workload, or thermal conditions.

5.3 Optical storage

Optical media, such as CDs, DVDs, and Blu-ray discs, use laser technology to read and write information. They are relatively inexpensive and can be useful for distributing software, media, or records. However, they have lower capacity and slower access than most modern SSDs and may require an optical drive that is no longer included in many computers. Long-term reliability depends on media quality, storage conditions, and handling.

5.4 Magnetic tape

Magnetic tape stores information sequentially. It is slow for locating an individual file but economical for very large backups and archives. Tape is portable, can be stored offline, and is relatively resistant to online cyberattacks when disconnected. Its main limitations are sequential access, specialized equipment, and the need for disciplined cataloguing and restoration procedures.

5.5 Network and cloud storage

Network-attached storage (NAS) provides files through a local network. It supports centralized management and collaboration but depends on network availability and appropriate access controls.

Cloud storage uses remote data-centre infrastructure accessed through a network. It can provide scalable capacity, synchronization, collaboration, and geographically distributed services. Cloud storage does not eliminate storage costs; organizations pay through subscriptions, data-transfer charges, administration, security controls, and dependence on the provider. Important questions include data location, contractual availability, privacy, encryption, backup responsibility, and the ability to migrate data to another provider.

5.6 Storage comparison

No storage option is universally best. A business may use SSDs for active databases, HDDs for economical capacity, cloud storage for collaboration, and tape or offline media for long-term backup.

6. Input Technologies

An input device converts human actions, physical conditions, or other data into signals that a computer can process.

6.1 Common human-input devices

The keyboard enters text, numbers, commands, and shortcuts. Keyboard design affects speed, accessibility, ergonomics, and error rates. The mouse, touchpad, and pointing devices control a cursor and support selection, dragging, and graphical interaction. A touchscreen combines display and input by detecting touch location and gestures. It is common in smartphones, kiosks, point-of-sale systems, and industrial terminals.

A microphone converts sound into digital data through a process involving sampling and encoding. A webcam captures still images or video. A scanner converts paper documents or images into digital form. Optical character recognition (OCR) can transform scanned text into editable characters, although accuracy may be affected by image quality, fonts, language, and layout.

6.2 Specialized input devices

A barcode reader identifies printed patterns used in retail, logistics, and inventory systems. A radio-frequency identification (RFID) reader communicates with tags using radio signals and may identify objects without direct line of sight. Biometric devices capture characteristics such as fingerprints, facial patterns, or iris features. Biometric input can improve convenience, but it raises significant privacy and security concerns because biometric characteristics cannot be changed as easily as passwords.

Sensors provide input from the physical environment. Temperature, pressure, motion, location, and health sensors support manufacturing, logistics, smart buildings, and Internet of Things applications. Sensor systems require calibration, data validation, secure communication, and decisions about retention and ownership.

7. Output Technologies

An output device converts processed digital information into a form that people or other systems can perceive or use.

A monitor presents text, graphics, and video. Important characteristics include screen size, resolution, refresh rate, brightness, colour accuracy, viewing angle, and accessibility. Resolution describes the number of addressable pixels; a higher resolution can display more detail but may require greater graphics capability.

A printer produces a physical representation of digital information. Inkjet printers are often suitable for colour images and low-volume use. Laser printers generally provide fast, sharp text output for office environments. Thermal printers are common for receipts, labels, and logistics documents. Printers involve continuing costs for ink, toner, paper, maintenance, energy, and secure disposal.

Speakers and headphones convert digital audio into sound. Projectors enlarge images for group viewing. Tactile and haptic devices provide touch-based feedback and can improve accessibility or support training simulations.

Output is not restricted to human senses. A computer may send output to another system through a network, an actuator, or a control device. For example, an inventory system may send a replenishment signal to an automated warehouse mechanism.

8. Peripheral Devices and Interfaces

A peripheral device is hardware connected to a computer that extends input, output, storage, communication, or specialized processing capabilities. Some peripherals are internal, such as expansion cards and internal drives; others are external, such as printers, cameras, keyboards, and removable drives.

Devices communicate through interfaces and protocols. USB supports many input, output, storage, and charging functions. HDMI and DisplayPort commonly transmit digital video and audio. Bluetooth supports short-range wireless connections, while Wi-Fi supports network communication. A device driver enables the operating system to communicate with particular hardware.

When selecting peripherals, organizations should consider compatibility, bandwidth, power requirements, security, accessibility, maintenance, and expected service life. Wireless convenience may introduce interference, battery management, pairing risks, and additional security requirements.

9. Applied Case: Storage and I/O for a Small Retailer

Background

This illustrative case concerns GreenBasket, a small retailer with three outlets. Each outlet uses point-of-sale terminals, barcode scanners, receipt printers, security cameras, and an inventory application. Managers also need access to sales reports from different locations.

Problem

The retailer stores transaction files on individual office computers. Reports are slow, backups are irregular, and a failed hard disk recently caused several days of disruption. Management must choose a more reliable arrangement without creating excessive cost or complexity.

Analysis

Active transaction processing requires fast access, so each terminal should use an SSD and sufficient RAM. Barcode scanners provide accurate and rapid input, while thermal printers are appropriate output devices for receipts. Centralized storage would simplify reporting, but a single local device would create a point of failure. Cloud services could support access across outlets, but the retailer must examine internet reliability, data protection, provider contracts, and recurring costs.

Approach and solution

GreenBasket adopts a hybrid design. Each terminal uses local SSD storage for the operating system and temporary operations. Sales data is transmitted securely to a central cloud-hosted application. A local NAS keeps short-term operational copies, while encrypted offline backups are created regularly and tested through sample restoration. Role-based access controls restrict staff access to only the functions required for their jobs. Barcode scanners reduce manual entry errors, and receipt printers are selected based on consumable cost and service availability.

Outcome

Reports become available across all outlets, terminal startup and application loading improve, and a damaged terminal can be replaced without losing central transaction data. The retailer still monitors provider availability, backup success, access logs, and recovery procedures.

Learning points

The case demonstrates that technology selection must match workload. Fast local storage supports active processing; centralized or cloud storage supports access and collaboration; offline backups improve resilience; and input–output devices should be selected according to transaction speed, accuracy, usability, and total cost rather than purchase price alone.

10. Reliability, Security, and Governance

Storage reliability is not the same as backup. Redundancy keeps services operating when a component fails, whereas a backup provides a separate copy that can be restored after accidental deletion, corruption, ransomware, or disaster. A useful operational principle is the 3-2-1 backup approach: maintain at least three copies, on two different types of media, with at least one copy offline or geographically separate. The precise design should reflect business requirements and regulatory obligations.

Organizations should classify information by sensitivity and retention requirements. Confidential data may require encryption at rest and in transit, strong authentication, access logging, and controlled deletion. Personal data should be collected only for legitimate purposes, retained no longer than necessary, and protected according to applicable privacy laws and institutional policies.

Storage and I/O devices also create environmental and ethical concerns. Data centres consume energy and water; electronic waste may contain hazardous materials; biometric systems can produce unequal error rates across demographic groups; and cloud services may transfer information across national borders. Responsible management includes lifecycle planning, secure disposal, accessibility testing, vendor due diligence, and transparent governance.

11. Contemporary Developments and Best Practices

Modern systems increasingly combine local, edge, and cloud resources. Edge computing processes data near the source, reducing latency and network traffic. This is useful for industrial sensors, autonomous systems, and real-time monitoring. AI applications intensify demand for fast storage and large memory because training and inference may process very large datasets. Nevertheless, AI does not remove the need for careful data governance: poor-quality, biased, or unlawfully collected data can produce harmful outcomes.

Best practices include:

1. Match storage technology to workload, not marketing claims.
2. Distinguish capacity from performance; a large device is not necessarily a fast device.
3. Use sufficient RAM to reduce dependence on virtual memory.
4. Maintain tested backups and documented recovery procedures.
5. Encrypt sensitive data and control physical and logical access.
6. Monitor device health, capacity, error rates, and replacement schedules.
7. Test compatibility and accessibility before large-scale deployment.
8. Calculate total cost of ownership, including maintenance, energy, training, downtime, and disposal.
9. Evaluate cloud providers for security, availability, portability, compliance, and exit arrangements.
10. Dispose of devices through secure data erasure and environmentally responsible recycling.

Chapter Summary

Computer memory and storage form a hierarchy balancing speed, capacity, permanence, and cost. Primary memory, especially RAM, holds active programs and data, while cache stores frequently used information close to the processor. Virtual memory extends the logical memory space by using secondary storage, but excessive reliance on it causes slow page transfers and thrashing. Secondary storage preserves information and includes HDDs, SSDs, optical media, magnetic tape, NAS, and cloud services.

Input devices convert human actions or environmental conditions into computer-readable signals. Output devices present results to people or communicate with other systems. Peripheral devices extend the capabilities of a computer through interfaces such as USB, Bluetooth, Wi-Fi, HDMI, and DisplayPort. Technology selection should consider capacity, speed, reliability, portability, cost, security, accessibility, and total cost of ownership. Effective organizational practice combines appropriate local storage, centralized or cloud services, tested backups, access controls, and responsible lifecycle management.

Key Terms

Primary memory: Memory directly accessible to the processor for active programs and data.

Cache memory: Small, high-speed memory that stores frequently used data and instructions.

Virtual memory: A technique that uses secondary storage to extend the logical memory available to programs.

RAM: Volatile random-access memory used for current processing.

ROM: Non-volatile memory commonly used for firmware.

Volatile memory: Memory that loses its contents when power is removed.

Non-volatile storage: Storage that retains information without continuous power.

Cache hit: A successful request for data already present in cache.

Cache miss: A request for data not present in cache.

Page fault: An event occurring when a required virtual-memory page is not in physical RAM.

Thrashing: Excessive movement of pages between RAM and storage that severely reduces performance.

HDD: Magnetic storage using rotating platters and moving read/write heads.

SSD: Flash-based storage with no moving mechanical parts.

Optical storage: Storage read or written using laser technology.

Peripheral device: Hardware that extends a computer’s input, output, storage, or communication functions.

Input device: Hardware that supplies data or commands to a computer.

Output device: Hardware that presents processed information.

Cloud storage: Remote storage provided through networked data-centre infrastructure.

Redundancy: Duplicate or alternative resources that help maintain service during failure.

Backup: A separate copy of data maintained for restoration after loss or damage.

Review Questions

Short Answer Questions

1. What is the difference between volatile and non-volatile memory?
2. Why is cache memory placed close to the processor?
3. What happens during a page fault?
4. State two advantages of SSDs over HDDs.
5. Give two examples each of input and output devices.

Long Answer Questions

1. Explain the computer memory hierarchy and discuss the trade-off between speed, capacity, and cost.
2. Compare primary memory, cache memory, virtual memory, and secondary storage.
3. Discuss the advantages and limitations of HDDs, SSDs, optical media, magnetic tape, NAS, and cloud storage.
4. Explain how input, output, and peripheral devices support business operations.
5. Discuss reliability, backup, security, privacy, and environmental issues in storage management.

Analytical / Application Questions

1. A small business needs fast application performance, inexpensive bulk storage, and protection against ransomware. Propose a storage arrangement and justify it using capacity, speed, reliability, portability, and cost.
2. An organization experiences slow performance whenever employees run many applications simultaneously. Analyze the possible roles of insufficient RAM, cache misses, and virtual memory.
3. Select suitable input and output technologies for a supermarket checkout system. Justify your choices in terms of accuracy, speed, usability, maintenance, and security.

Multiple Choice Questions with Answers

Question 1

Which memory normally holds programs and data that are currently being processed?

A. RAM

B. Optical disc

C. Magnetic tape

D. Printer memory only

Correct Answer: A. RAM

Question 2

What is the main purpose of cache memory?

A. To provide permanent file archiving

B. To store frequently used data close to the processor

C. To replace all secondary storage

D. To convert printed pages into text

Correct Answer: B. To store frequently used data close to the processor

Question 3

What is a likely consequence of frequent page faults?

A. Improved printer resolution

B. Thrashing and reduced performance

C. Permanent loss of all firmware

D. Increased optical-disc capacity

Correct Answer: B. Thrashing and reduced performance

Question 4

Which storage device uses rotating magnetic platters?

A. SSD

B. HDD

C. Blu-ray disc

D. USB keyboard

Correct Answer: B. HDD

Question 5

Which characteristic is generally associated with an SSD rather than an HDD?

A. Moving read/write heads

B. Rotating platters

C. Faster access and greater shock resistance

D. Sequential-only access

Correct Answer: C. Faster access and greater shock resistance

Question 6

Which device is primarily used to capture printed documents digitally?

A. Scanner

B. Projector

C. Speaker

D. Monitor

Correct Answer: A. Scanner

Question 7

Which device is an output device?

A. Barcode reader

B. Microphone

C. Thermal receipt printer

D. Temperature sensor

Correct Answer: C. Thermal receipt printer

Question 8

A cache has a hit ratio of 0.90, cache access time of 2 ns, and miss-handling time of 50 ns. What is the simplified average access time?

A. 2.0 ns

B. 5.0 ns

C. 6.8 ns

D. 52.0 ns

Correct Answer: C. 6.8 ns

Question 9

Which storage option is especially suitable for large offline archives because it is inexpensive per unit of capacity?

A. Magnetic tape

B. Processor register

C. L1 cache

D. Keyboard buffer

Correct Answer: A. Magnetic tape

Question 10

Which statement best describes a backup?

A. A duplicate or separate copy used for restoration after data loss

B. A faster type of processor cache

C. A device that displays output

D. A method of increasing screen resolution

Correct Answer: A. A duplicate or separate copy used for restoration after data loss