Skip to topic
    ← Back to course topics

    Secondary storage devices — AQA A-Level Computer Science

    Secondary storage devices provide non-volatile, long-term data retention for software and files when a computer is powered off. For AQA A-Level Computer Science, you must understand the operational physics, comparative characteristics, and trade-offs of magnetic, optical, and solid-state storage technologies.

    Secondary storage devices explained

    Primary storage, such as RAM, is fast but volatile: its contents are lost when power is removed, and its capacity is limited and relatively expensive per gigabyte.

    Read the full explanation

    Secondary storage fills this gap. It is non-volatile, so it retains data and programs when the computer is switched off, and it offers much larger capacities at lower cost per gigabyte. A computer needs secondary storage to hold the operating system, applications, user files such as documents, photos and databases, and data that must survive between sessions. When a program is run, the processor fetches it from secondary storage and loads it into RAM, where it can be accessed quickly. Explaining the need therefore means linking volatility, capacity, cost and persistence to the role of storing programs and data long term.

    Your focus

    1. Explain the need for secondary storage within a computer system.

    Which command word is this?

    Set out the technical causes, working mechanisms, or reasons behind a storage device's operation. Requires precise technical terms (e.g., 'tunnelling', 'actuator arm', 'constructive interference').

    Secondary storage devices exam tips

    Topic Overview

    Secondary storage encompasses non-volatile hardware devices that retain data, operating system files, and software applications independently of continuous electrical power. Unlike primary memory (RAM), which is volatile and directly accessible by the central processing unit, secondary storage handles long-term storage across high capacities at significantly lower costs per gigabyte.

    Understanding the physical mechanisms of magnetic disks, optical discs, and solid-state NAND flash is essential for analysing system performance. Selecting appropriate secondary storage requires evaluating trade-offs among read/write speed, cost per unit capacity, physical footprint, durability, and power consumption within computing systems.

    Key Concepts
    • →Non-volatility: Secondary storage retains its stored binary states when power is removed, whereas RAM loses its contents.
    • →Magnetic storage mechanisms: Hard disk drives (HDDs) utilise magnetised ferromagnetic surfaces arranged in concentric tracks divided into sectors, read and written by electromagnetic heads.
    • →Optical storage mechanisms: Media such as CDs, DVDs, and Blu-rays use a laser to detect phase transitions between microscopic pits and lands along a continuous spiral track.
    • →Solid-state flash memory: SSDs utilise arrays of non-volatile NAND flash cells, trapping electrons in floating gates or charge traps via Fowler-Nordheim tunnelling to modify threshold voltages.
    • →Comparative metrics: Evaluation criteria include latency (seek time and rotational delay), throughput, cost per gigabyte, physical durability, portability, and capacity.
    Marking Points
    • Identifies that secondary storage is non-volatile, so data persists when power is removed.
    • Explains that secondary storage provides larger capacity than primary storage at lower cost per unit.
    • States that the operating system, applications and user data are stored on secondary storage.
    • Explains that programs are copied from secondary storage into RAM before the processor can execute them.
    • Links the need for secondary storage to the limited size and volatility of RAM.
    Examiner Tips
    • 💡Structure answers as a contrast with primary storage, covering volatility, capacity and cost.
    • 💡Use the term non-volatile precisely and explain its consequence for data retention.
    • 💡Give a concrete example such as saving a document or installing an application to show the need in practice.
    • 💡Break down HDD access time into its distinct technical components: seek time (moving the actuator arm across tracks) and rotational latency (waiting for the correct sector to spin beneath the head).
    • 💡When answering comparative 6-mark questions, always relate storage characteristics directly to the stated scenario context (e.g., power consumption and vibration resistance for a wearable device versus cost per terabyte for a cloud backup server).
    • 💡Use standard data unit terminology correctly: differentiate clearly between mebibytes (base 2, MiB) and megabytes (base 10, MB) if storage capacity calculations appear.
    Common Mistakes
    • Suggests that secondary storage is used for the processor's immediate working data; correct by stating that this is the role of RAM and registers.
    • Describes secondary storage as faster than RAM; correct by explaining that it is slower but holds far more data permanently.
    • Gives a list of devices without explaining why storage must be persistent and high capacity; correct by stating the functional reason for each.
    • Believing that an SSD can overwrite data in-place like an HDD: SSDs can read and write data in small pages (typically 4KB), but they can only erase data in entire blocks (typically 128 to 256 pages). An existing page must be erased before being rewritten, requiring read-modify-write cycles and wear levelling.
    • Assuming pits represent binary 0 and lands represent binary 1 on optical media: Pits and lands themselves do not directly map to binary bits. Instead, the transition edge from a pit to a land (or vice versa) scatters light and represents a 1, while the flat continuity of a pit or land represents a 0.
    • Thinking secondary storage is directly accessible by the CPU: The CPU can only fetch and execute instructions directly from primary storage (registers, cache, and RAM). Secondary storage data must first be loaded into primary memory via the I/O bus before execution.
    Revision Plan
    1. 1Day 1-2: Master the physical operation and mechanisms of magnetic HDDs, optical media (CD/DVD/Blu-ray), and solid-state flash memory.
    2. 2Day 3-4: Build a comparative matrix covering the 6 core criteria: Capacity, Speed, Portability, Durability, Reliability, and Cost.
    3. 3Day 5-6: Practice exam-style scenario questions requiring technology recommendations and detailed technical justifications.
    4. 4Day 7: Review active recall flashcards on optical disc transitions and NAND floating gate tunnelling.
    Exam Question Types
    • 📋Comparative Scenario Evaluation (6-9 marks): Given a specific practical scenario (e.g., digital camera, server farm, mobile console), evaluate two or three storage media types and justify the best choice.
    • 📋Technical Explanation (2-4 marks): Explain the physical mechanism behind reading or writing data for a specific medium (such as laser reflection off pits/lands or electron trapping in flash cells).
    • 📋Data Calculation and Capacity Calculations (2-3 marks): Calculate total transfer times or determine storage capacity requirements based on file sizes and bit rates.
    Command Word Expectations (AQA)
    Explain

    Set out the technical causes, working mechanisms, or reasons behind a storage device's operation. Requires precise technical terms (e.g., 'tunnelling', 'actuator arm', 'constructive interference').

    Discuss / Evaluate

    Explore multiple perspectives or technologies, weighing strengths and limitations in relation to a given scenario, concluding with a supported and justified judgment.

    Compare

    Identify specific similarities and differences between two or more storage technologies across defined metrics like speed, cost, and durability, rather than describing each in isolation.

    How Students Lose Marks (Examiner Pitfalls)
    Pitfall: Confusing how data is read on an optical disc by stating that 'pits represent 0s and lands represent 1s'.
    ❌ Weak Answer (Loses Marks):A laser reads the bumps on the disc where pits are 0s and lands are 1s to make binary code.
    Example improved answer:A laser beam is focused onto the spiral track of the disc. When the laser hits a flat surface (land) or the bottom of a pit, light is reflected straight back into a photodiode sensor. However, when the laser passes over the edge between a pit and a land, the beam is scattered by destructive interference. The sensor detects this transition as a change in reflection, representing a binary 1, whereas continuous land or pit bottom represents binary 0.
    Examiner Tip: Always refer to transitions between lands and pits as binary 1s, and the absence of a transition as binary 0. Never claim pits equal 0 and lands equal 1.
    Pitfall: Vague comparisons between SSDs and HDDs that rely on generic terms like 'better' or 'faster' without technical justification.
    ❌ Weak Answer (Loses Marks):SSDs are better than HDDs because they are newer, faster at loading games, and do not break as easily.
    Example improved answer:Solid-state drives offer substantially lower latency and higher random read/write throughput than hard disk drives because they contain no moving parts, eliminating rotational delay and head seek time. Furthermore, SSDs consume less power and possess superior mechanical shock resistance due to using NAND flash memory arrays with floating-gate or charge-trap transistors.
    Examiner Tip: Use technical metrics: mention 'latency', 'seek time', 'rotational delay', 'durability', and 'power consumption' instead of informal adjectives.
    Step-by-Step Worked Solutions

    Question: A wildlife researcher needs an autonomous field recording device to record high-definition video continuously for 7 days in sub-zero temperatures. Recommend and justify the most suitable type of secondary storage technology (magnetic, optical, or solid-state) and explain why the other two options are unsuitable.

    1. 1.Step 1: Identify key operational requirements: continuous video writing (high capacity and write endurance), portable field equipment exposed to physical movement and sub-zero temperatures, and energy efficiency for battery power.
    2. 2.Step 2: Evaluate Solid-State Storage (SSD): SSDs have no moving mechanical parts, providing high shock resistance and reliable operation across wider temperature ranges. They offer silent operation and low power consumption, making them highly suitable for battery-powered remote equipment.
    3. 3.Step 3: Evaluate and reject Magnetic Storage (HDD): Hard disks contain mechanical platters and read/write heads that can suffer head crashes if jostled in the field. Lubricants in mechanical bearings can also seize or thicken in sub-zero temperatures, leading to drive failure.
    4. 4.Step 4: Evaluate and reject Optical Storage (CD/DVD/Blu-ray): Optical media has insufficient capacity for 7 days of continuous HD video, very slow write speeds, and mechanical laser trays vulnerable to dust, moisture, and vibration.
    5. 5.Step 5: Synthesise recommendation: Solid-state storage is the superior choice due to environmental durability, energy efficiency, and lack of moving parts.
    Final Answer: Solid-state storage is the most suitable technology because it lacks moving parts, providing exceptional shock resistance, operational reliability in sub-zero conditions, and low power draw. Magnetic HDDs risk head crashes and bearing freeze, while optical media lacks the required capacity and write throughput.

    Question: Explain the internal operation of NAND flash memory used in solid-state drives during read and write operations, referencing floating-gate transistors and electron movement.

    1. 1.Step 1: Describe the structure: NAND flash memory consists of an array of cells, each containing a MOSFET with a control gate and a floating gate separated by insulating oxide layers.
    2. 2.Step 2: Explain writing (programming): A high positive voltage is applied to the control gate. Electrons are pulled from the p-substrate channel and tunnel through the insulating oxide barrier into the floating gate via Fowler-Nordheim tunnelling, becoming trapped.
    3. 3.Step 3: Explain the state determination: Trapped negative electrons increase the threshold voltage needed to turn on the transistor. When an intermediate read voltage is applied to the control gate, a cell with trapped charge does not conduct (sensed as a binary 0), whereas a cell with no trapped charge conducts current (sensed as a binary 1).
    4. 4.Step 4: Explain erasing: High negative voltage across the control gate or high positive voltage to the substrate pulls electrons back out of the floating gate, resetting the block to 1s.
    Final Answer: NAND flash relies on floating-gate transistors. Electrons are forced into the floating gate via quantum tunnelling by applying a high voltage to the control gate. Trapped electrons alter the transistor's threshold voltage; during a read, current flow is blocked (binary 0) or allowed (binary 1). Blocks are erased by reversing voltage to discharge trapped electrons.
    Active Recall Memory Test
    What physical feature on an optical disc indicates a binary 1 during a read operation?
    Key Fact: A transition (edge) between a pit and a land, which causes destructive interference and scatters the reflected laser light.
    Which two physical delays make up the primary access latency of a hard disk drive?
    Key Fact: Seek time (moving the read/write head to the correct track) and rotational latency (waiting for the target sector to rotate beneath the head).
    Why must an SSD erase data in blocks rather than individual pages?
    Key Fact: Due to the electrical architecture of NAND flash, high erase voltages cannot be targeted at individual pages without causing field interference across adjacent cells, requiring entire blocks to be cleared at once.
    How does electron leakage affect solid-state storage left unpowered for extended periods?
    Key Fact: Electrons trapped in the floating gates can slowly tunnel back through the insulating oxide layer over time, degrading the stored threshold voltage and causing data corruption (bit rot).
    Frequently Asked Questions
    Why cannot the CPU directly execute programs from an SSD or HDD?
    The CPU's program counter and instruction registers communicate over high-speed system buses that only interface with primary memory (RAM and cache). Secondary storage devices are connected via slower I/O controllers and buses (such as SATA or NVMe PCIe). Operating systems must load program instructions from secondary storage into RAM before the CPU can fetch, decode, and execute them.
    What is wear levelling in solid-state drives and why is it necessary?
    Each NAND flash memory cell can only withstand a finite number of program-erase cycles before the insulating oxide layer degrades and fails to trap electrons reliably. Wear-levelling algorithms in the SSD controller distribute write and erase operations evenly across all physical blocks in the drive. This prevents specific blocks from wearing out prematurely due to repeated writes to the same logical address.
    How do CD, DVD, and Blu-ray discs differ in physical capacity?
    All three use optical storage, but Blu-ray discs have a much higher capacity (25GB+) than DVDs (4.7GB) and CDs (700MB). This is achieved by using a laser with a shorter wavelength (405nm blue-violet laser for Blu-ray versus 650nm red for DVD and 780nm infrared for CD). The shorter wavelength allows the beam to focus to a much smaller diffraction-limited spot, enabling tracks and pits to be packed significantly closer together.
    What is the difference between latency and throughput in secondary storage?
    Latency is the initial delay between requesting data and receiving the first byte, which in HDDs is dominated by mechanical seek time and rotational delay. Throughput (or transfer rate) is the sustained speed at which consecutive data bytes can be streamed once reading or writing has commenced. SSDs have near-zero mechanical latency and high throughput, while HDDs have high latency but respectable sequential throughput.
    Is magnetic storage still relevant now that SSDs are common?
    Yes, magnetic storage remains critically important in enterprise computing, enterprise backup systems, and hyperscale data centres. Hard disk drives provide a significantly lower cost per gigabyte than SSDs for storing high volumes of sequential data, while magnetic tapes remain the industry standard for cold archival storage due to unmatched capacity per unit cost and zero idle power requirements.