Secondary storage, compression and cloud computing – IB Computer Science exam questions with mark scheme

1Core6 marks

For each scenario, state the most suitable type of secondary storage.

(a) The built-in storage of a low-cost smartphone. [1]

(b) Removable storage for a helmet-mounted action camera that records video. [1]

(c) A small architecture practice where eight desktop computers must all access the same project files over the office network. [1]

(d) A charity posting a read-only copy of a training video to volunteers who have no internet access. [1]

(e) Outline two reasons why the storage in (a) is used instead of a separate internal SSD. [2]

(a) eMMC (embedded MultiMediaCard) (accept embedded flash memory soldered to the board; reject 'flash' alone, reject memory card)(1)
(b) Memory card, e.g. microSD / SD card (accept flash memory card)(1)
(c) NAS (network-attached storage) (accept a file server; reject cloud storage since the question specifies the office network)(1)
(d) Optical disc, e.g. DVD / Blu-ray (accept CD-R; do not accept USB flash drive or external HDD, which are rewritable and costlier to post in bulk)(1)
eMMC is cheaper (per device) than a full SSD with its own controller/interface, keeping the phone's price low(1)
eMMC is soldered directly onto the motherboard so it takes up very little space / is very compact and low-power (accept: no separate drive enclosure or connector needed)(1)
Answer: (a) eMMC; (b) memory card (microSD); (c) NAS; (d) optical disc (DVD/Blu-ray); (e) cheaper, and compact/soldered to board/low power.
Examiner tip: Name the specific device (eMMC, NAS, Blu-ray) rather than 'flash memory' or 'external hard drive' – generic answers rarely score.

2Core4 marks

(a) Distinguish between lossy and lossless compression. [2]

(b) A software company compresses an installer for its application before customers download it. State whether lossy or lossless compression must be used, giving a reason. [1]

(c) A radio station compresses podcast episodes for streaming to mobile phones. State whether lossy or lossless compression is more appropriate, giving a reason. [1]

Lossy compression permanently removes/discards some data (usually detail people are unlikely to notice) so the original cannot be reconstructed exactly(1)
Lossless compression reduces file size without losing any data, so the original file can be reconstructed exactly / bit-for-bit (accept: lossy usually gives a smaller file than lossless as a comparison point only if one of the two definitions is given)(1)
(b) Lossless, because every bit of program code must be restored exactly or the program would not run correctly / would be corrupted(1)
(c) Lossy, because a much smaller file streams faster / uses less mobile data and the small loss in audio quality is barely noticeable to listeners (reason required for the mark)(1)
Answer: (a) Lossy discards data permanently; lossless allows exact reconstruction. (b) Lossless – code must be exact. (c) Lossy – smaller file, quality loss acceptable.
Examiner tip: 'Lossy loses quality' is not enough: say the data is removed permanently and cannot be recovered.

3Standard4 marks

The diagram shows one row of 20 pixels from a black-and-white image. Uncompressed, each pixel is stored in 1 byte.

The image is compressed using run-length encoding (RLE). Each run is stored as a pair: a count followed by a colour, e.g. 6W. Each pair takes 2 bytes (1 byte for the count, 1 byte for the colour).

(a) Write the run-length encoding of this row. [2]

(b) Calculate the percentage reduction in the storage needed for this row when it is compressed with RLE. Show your working. [2]

(a) At least three runs correct and in order (e.g. 6W 3B ...)(1)
(a) Fully correct: 6W 3B 2W 8B 1W (accept any clear separator, or W6 B3 ... if consistent)(1)
(b) Uncompressed = 20 bytes and compressed = 5 pairs × 2 = 10 bytes (follow through from their number of runs in (a))(1)
(b) Reduction = (20 − 10) / 20 × 100 = 50% (follow through)(1)
Answer: (a) 6W 3B 2W 8B 1W; (b) 20 bytes → 10 bytes, 50% reduction.
Examiner tip: Count the pixels in each run carefully and check your counts add up to 20 before calculating.

Stuck on Secondary storage, compression and cloud computing?

Get it sorted in a free 30-minute session

We work through your weakest questions on this topic on Zoom. I reply the same day.

Free 30-min session on this topic Send me 5 questions a day

4Standard4 marks

A text compression program uses run-length encoding in which each run is written as a count followed by the character, e.g. 3X means XXX.

(a) Decode the following: 5A2B1C4A [1]

(b) The string ABCDABCD is encoded with the same program. Explain why the encoded version is larger than the original. [2]

(a) AAAAABBCAAAA (12 characters)(1)
(b) There are no repeated consecutive characters, so every run has length 1(1)
(b) Each single character is therefore stored as two characters (count + character), e.g. 1A1B1C1D1A1B1C1D = 16 characters vs 8, so the size doubles (accept 'it doubles' if the reason for the doubling is clear)(1)
(c) Lossless, because decoding reproduces the original data exactly / no data is discarded(1)
Answer: (a) AAAAABBCAAAA; (b) no runs, so each character becomes 1+character, 16 vs 8 characters; (c) lossless – exact reconstruction.
Examiner tip: RLE only helps when data contains long runs of identical values; always say why the size grows, not just that it does.

5Standard5 marks

A student is choosing internal storage for a new laptop. The table shows two options.

Drive	Capacity	Sequential read speed	Price
SSD	512 GB	3500 MB/s	£45
HDD	2 TB	160 MB/s	£50

(a) A game loads a 7000 MB data file at start-up. Calculate how many seconds faster the file is read from the SSD than from the HDD. [2]

(b) Apart from speed, explain two reasons why an SSD is more suitable than an HDD for a laptop. [2]

(a) SSD time = 7000/3500 = 2 s and HDD time = 7000/160 = 43.75 s (both needed)(1)
(a) Difference = 41.75 s (accept 41.8 s or 42 s)(1)
(b) No moving parts, so it is more resistant to damage from knocks/drops when the laptop is carried around(1)
(b) Uses less power (longer battery life) / is lighter, thinner or silent / produces less heat (any one distinct reason other than durability)(1)
(c) When large capacity at low cost per GB matters more than speed, e.g. storing a large video/photo library or backups (2 TB for about the same price) (accept a desktop or external backup drive used for bulk storage)(1)
Answer: (a) 2 s vs 43.75 s, so 41.75 s faster; (b) no moving parts (shock-resistant), lower power/lighter/silent; (c) bulk storage of large files where cost per GB matters.
Examiner tip: In (b) 'faster' is excluded by the question – choose physical reasons linked to a portable device.

6Core4 marks

For each of the following, identify the cloud service model being used (SaaS, PaaS or IaaS).

(a) A school's staff write reports and send email using applications that run in a web browser, without installing anything. [1]

(b) A developer uploads code for a web app; the provider supplies the operating system, runtime environment and database engine and scales the app automatically. [1]

(c) A research group rents virtual machines, installs its own choice of Linux distribution and configures its own simulation software. [1]

(d) State which of the three models gives the customer the greatest control over the computing environment. [1]

(a) SaaS (Software as a Service)(1)
(b) PaaS (Platform as a Service)(1)
(c) IaaS (Infrastructure as a Service)(1)
(d) IaaS (follow through not allowed; reject SaaS/PaaS)(1)
Answer: (a) SaaS; (b) PaaS; (c) IaaS; (d) IaaS.
Examiner tip: Ask 'what does the customer manage?': only the data (SaaS), the code (PaaS), or the OS and everything above it (IaaS).

7Hard5 marks

The diagram shows which layers of a cloud service are managed by the customer and which are managed by the cloud provider for a service model labelled X.

(a) Identify service model X. [1]

A games start-up runs its multiplayer servers on X. It now needs a database engine version and a network driver that X does not support.

(b) Explain why moving to IaaS would solve this problem. [2]

(a) PaaS (Platform as a Service)(1)
(b) With IaaS the customer manages the operating system, middleware and runtime (only the hardware/virtualisation is managed by the provider)(1)
(b) so the start-up can install/configure any OS, driver or database version it chooses, giving greater control/flexibility(1)
(c) The start-up becomes responsible for installing, patching/updating, securing and maintaining the OS and software layers(1)
(c) which needs more technical staff/time/cost, and mistakes (e.g. unpatched OS) could cause downtime or security vulnerabilities (accept: loses automatic scaling, so must configure scaling itself)(1)
Answer: (a) PaaS; (b) IaaS hands the OS/middleware/runtime layers to the customer, so it can install the version/driver it needs; (c) more management responsibility – greater cost/skills, risk of downtime or vulnerabilities.
Examiner tip: Link control to responsibility: more control in IaaS always means more for the customer to maintain.

8Hard5 marks

Photographs are often compressed using transform coding (for example, in the JPEG format).

(a) Describe how transform coding reduces the size of an image file. [4]

(b) A photo is opened, edited and saved as a JPEG many times. Explain why its quality gets steadily worse. [1]

(a) The image is divided into small blocks of pixels (e.g. 8 × 8)(1)
(a) A mathematical transform (e.g. discrete cosine transform) converts each block from pixel values into frequency components/coefficients(1)
(a) High-frequency components (fine detail the human eye is less sensitive to) are quantised/rounded so many become zero / are discarded (this is the lossy step)(1)
(a) The remaining coefficients, which contain many zeros/repeated values, are then encoded efficiently, e.g. with run-length or Huffman encoding(1)
(b) Each save applies the lossy step again, so more data is discarded each time and lost data cannot be recovered (accept 'generation loss' with explanation)(1)
Answer: (a) Blocks → transform to frequency coefficients → quantise away fine detail (many zeros) → encode the remaining values (e.g. RLE). (b) Lossy compression repeated each save; discarded data is never recovered.
Examiner tip: Mention both the transform into frequencies and the discarding/quantising of detail – the transform alone does not shrink the file.

9Hard5 marks

A camera takes photographs of 4000 × 3000 pixels using 24-bit colour. Before uploading to a cloud storage service, each photo is compressed with a compression ratio of 12 : 1.

(a) Calculate the size of one uncompressed photograph in megabytes (1 MB = 10⁶ bytes). Show your working. [2]

(b) Calculate the size of the compressed photograph in MB. [1]

(c) The upload speed is 20 Mbit/s (1 Mbit = 10⁶ bits). Calculate how many seconds are saved per photograph by uploading the compressed version instead of the uncompressed one. [2]

(a) 4000 × 3000 × 24 = 288 000 000 bits or 4000 × 3000 × 3 = 36 000 000 bytes (method)(1)
(a) 36 MB(1)
(b) 36 / 12 = 3 MB (follow through from (a))(1)
(c) Converts to bits and divides by speed: uncompressed 288 Mbit / 20 = 14.4 s and compressed 24 Mbit / 20 = 1.2 s (allow the direct route: (36 − 3) MB × 8 = 264 Mbit, ÷ 20) (follow through)(1)
(c) Time saved = 13.2 s (follow through; if bytes not converted to bits, answer 1.65 s scores max 1 of the 2 marks for (c))(1)
Answer: (a) 36 MB; (b) 3 MB; (c) 14.4 s − 1.2 s = 13.2 s saved.
Examiner tip: Storage is in bytes but network speeds are in bits – multiply by 8 before dividing by the upload speed.

10Stretch6 marks

A secondary school stores students' coursework on a network-attached storage (NAS) device in its server room. The school is considering replacing the NAS with a cloud storage service (SaaS) that students can access from home.

Evaluate the proposal to replace the NAS with cloud storage. [6]

Cloud: accessible from any internet-connected device/location, so students can work on coursework at home(1)
Cloud: provider handles hardware maintenance, backups/redundancy and upgrades, reducing the school's need for technical staff(1)
Cloud: scalable – storage can be increased as needed without buying new drives; pay-as-you-go/subscription instead of upfront hardware cost(1)
Cloud: depends on a reliable internet connection; outages or slow bandwidth stop access, while a NAS still works on the local network(1)
Cloud: ongoing subscription costs may exceed the one-off cost of a NAS over several years(1)
Cloud: student data is held by a third party, possibly in another country – privacy/data protection (e.g. GDPR) and security concerns; less control than an on-site NAS(1)
NAS: fast local transfer speeds and full control, but a single point of failure (theft, fire, drive failure) unless backed up off-site(1)
A justified conclusion, e.g. move to cloud for access and resilience while checking data-protection compliance, or a hybrid with NAS plus cloud backup(1)

Level	Marks	What it looks like
Level 3	5–6	Balanced evaluation with developed advantages and disadvantages of both cloud storage and the NAS, consistently applied to the school; reaches a justified conclusion.
Level 2	3–4	Several relevant points covering both benefits and drawbacks, with some explanation and application to the school; conclusion absent or weakly justified.
Level 1	1–2	Limited, mostly one-sided or generic points about cloud storage or NAS with little explanation.
Level 0	0	No relevant content.

Answer: Balanced evaluation: cloud gives remote access, managed backups and scalability; NAS gives local speed, control and no internet dependence or recurring fee; justified conclusion.
Examiner tip: 'Evaluate' needs both sides and a conclusion tied to the school's context – a list of cloud advantages caps you in the middle band.

Stuck on this topic?

Book a free 30-minute session on WhatsApp and we'll find exactly which step is costing you the marks. Or build a timed paper on it.

Free 30-minute session