Why is "Using 'serial' and 'sequential' interchangeably, or saying a sequential file is just stored 'in order'." a common mistake in File organisation and access?

Why it happens: Both words sound similar and both involve order, so students do not realise serial = arrival order while sequential = key-field order. How to avoid it: Always state the distinguishing fact: serial = order of ARRIVAL (no key); sequential = order of a KEY FIELD. Mention the key field explicitly. Source: Examiner reports repeatedly note candidates confusing serial and sequential organisation.

Why is "Claiming a serial file allows direct access to a record." a common mistake in File organisation and access?

Why it happens: Students assume any file can jump to a record, forgetting a serial file has no key order and no computed address. How to avoid it: Remember: serial and sequential files support sequential access; sequential and random files support direct access. A serial file offers sequential access only.

Why is "Numbering a table of N slots from 1 to N (or computing key MOD N but expecting addresses 1..N)." a common mistake in File organisation and access?

Why it happens: Students are used to counting from 1, but MOD N produces remainders 0 to N−1. How to avoid it: A table of 10 slots uses addresses 0 to 9. key MOD 10 can only ever give 0–9, so start the table at slot 0.

Why is "Confusing MOD (remainder) with integer division, e.g. writing 47 MOD 10 = 4." a common mistake in File organisation and access?

Why it happens: DIV and MOD are easily swapped; 47 DIV 10 = 4 but 47 MOD 10 = 7. How to avoid it: MOD gives the REMAINDER after dividing. For MOD 10 it is simply the last digit: 47 MOD 10 = 7, 53 MOD 10 = 3.

Why is "Describing a collision as 'when the file is full' or 'when there is no space'." a common mistake in File organisation and access?

Why it happens: Students associate the problem with running out of room rather than the precise hashing cause. How to avoid it: Define a collision precisely: when the hashing algorithm produces the SAME address for two DIFFERENT keys. It can happen even in a mostly empty table.

Why is "Explaining how a collision is STORED (next free slot / overflow area) but never how the record is found again." a common mistake in File organisation and access?

Why it happens: Students focus on writing the record and forget reading is the harder half. How to avoid it: Always finish the answer with retrieval: re-hash the key to its home address, then search forward (open addressing) or search the overflow area until the matching key is found.

Cambridge International A Levels Computer Science (9618)

File organisation and access

0% Complete

Detailed Study Notes

Detailed notes on Data Representation for Cambridge International A Levels Computer Science, covering key concepts, explanations, examples, and exam-focused revision points.

File Organisation and Access — Cambridge AS & A Level Computer Science 9618 Study Notes (2027-2029)

The three ways records are organised in a file (serial, sequential by a key field, random by a record key), the two ways they are accessed (sequential access and direct access), and how a hashing algorithm turns a record key into a storage address — including collisions and how to resolve them. This is A-Level content examined in Paper 3, written with no calculator, so every hashing example here uses small, hand-friendly numbers.

At a glance

A serial file stores records in order of arrival (no key order) — new records are appended.
A sequential file stores records in order of a key field (e.g. ascending account number).
A random file computes each record's position from its record key using a hashing algorithm.
Sequential access = read records in order from the start; used by serial and sequential files.
Direct access = jump straight to a record; available for sequential and random files.
A hashing algorithm maps a key to an address, e.g. address = key MOD numberOfSlots.
A collision is when two different keys hash to the same address.
Resolve collisions with open addressing (next free slot) or an overflow area (separate region).
Choose the organisation by the access pattern: append/read-all → serial; process-all-in-order → sequential; fast single-record lookup → random.

What you’ll learn

Mapped to the Cambridge International A Level 9618 syllabus (2027-2029).

13.2 — Show understanding of methods of file organisation: serial, sequential (using a key field) and random (using a record key).
13.2 — Select an appropriate method of file organisation for a given problem.
13.2 — Show understanding of methods of file access: sequential access for serial and sequential files; direct access for sequential and random files.
13.2 — Describe and use different hashing algorithms to read data from, and write data to, a random or sequential file.
13.2 — Show understanding of collisions and how they are resolved (e.g. next free slot / overflow area).

Records, fields and the key

A file is a collection of records; each record is made of fields, and one field is usually the key.

A file stores many records. Each record is made of fields — for example a customer record has the fields customerID, name, balance.

One field is special: the key field. Its value uniquely identifies the record (no two records share it) and it is the value used to order records or to calculate where to store them.

Term	Meaning
Record	one complete entry (e.g. one customer)
Field	one item of data within a record (e.g. name)
Key field / record key	the field that uniquely identifies a record (e.g. customerID)

The three file organisations below differ only in how the records are arranged, and that in turn decides how they can be accessed.

File → records → fields.
The key field uniquely identifies a record.
The key is used to order records (sequential) or to compute an address (random).

Three methods of file organisation

Serial = arrival order; sequential = key-field order; random = address computed from the key.

Serial file. Records are stored one after another in the order they arrive (chronological order). There is no key ordering — a new record is simply appended to the end. Good for logs and audit trails.

Sequential file. Records are stored in order of a key field (e.g. ascending employee number). They are sorted, so they are ideal for processing every record once, in order (batch processing like a payroll).

Random (direct) file. The storage position (address) of each record is calculated from its record key by a hashing algorithm. This lets the system jump straight to a single record — perfect when you must fetch one record fast from a very large file.

The diagram contrasts serial (arrival order) with sequential (key order):

Sequential (key order) key 12 key 34 key 57 key 90 sorted by key field

Same four records, organised two different ways — the serial file keeps them in the order they came in, the sequential file sorts them by key.

Serial: arrival order, no key ordering, append to the end.
Sequential: ordered by a key field (sorted).
Random: address computed from the record key by hashing.

See the full worked example for file organisation and access →

Two methods of file access

Sequential access reads in order; direct access jumps to the record.

Access is about how you get to a record, which depends on the organisation.

Sequential access — read records one after another from the start of the file until you reach the one you want (or the end). To reach the 100th record you must pass the 99 before it. This is the access method for serial files (you have no choice) and is also used to read sequential files in order.

Direct access (also called random access) — go straight to the required record without reading the ones before it, because its position is known or can be computed. Available for:

Sequential files — because they are ordered by key, you can jump in (e.g. via an index) rather than read everything.
Random files — the address is computed straight from the key by hashing.

File organisation	Sequential access?	Direct access?
Serial	✅ Yes (only option)	❌ No
Sequential	✅ Yes	✅ Yes
Random	— (used for whole-file reads)	✅ Yes

The headline rule to memorise: serial + sequential files use sequential access; sequential + random files allow direct access.

Sequential access = read in order from the start.
Direct access = jump straight to the record.
Serial files: sequential access only.
Sequential and random files: direct access available.

See the full worked example for file organisation and access →

Hashing: turning a key into an address

A hashing algorithm calculates a storage address from the record key.

A random file needs a rule that says where to put each record. That rule is the hashing algorithm: it takes the record key and performs a calculation to produce a storage address.

A simple, exam-standard hashing algorithm is:

$\text{address} = \text{recordKey} \bmod \text{numberOfSlots}$

MOD gives the remainder after division. For a 10-slot table (addresses 0–9), key MOD 10 is simply the last digit of the key.

47 MOD 10 = 7 53 MOD 10 = 3 20 MOD 10 = 0

slot 0 ← key 20 slot 1 ·· slot 2 slot 3 ← key 53 slot 4 ·· slot 6 slot 7 ← key 47 slot 8 ·· slot 9 10 slots: addresses 0–9

Reading a record back uses the same hash: to find key 47 the system computes 47 MOD 10 = 7 and looks in slot 7. Because the same key always hashes to the same address, writing and reading both use the one calculation.

Other hashing algorithms exist (e.g. add up the digits of the key, then MOD the table size; or take the middle digits), but they all share the same purpose — convert a key into an address.

Hashing algorithm: key → storage address.
Standard form: address = key MOD numberOfSlots.
MOD = remainder; key MOD 10 = the last digit.
Reading uses the same hash as writing.

See the full worked example for file organisation and access →

Collisions and how to resolve them

Two keys hashing to the same address collide; resolve with the next free slot or an overflow area.

A hashing algorithm cannot guarantee a different address for every key. When two different keys produce the same address, that is a collision — both records 'want' the same slot, but it can only hold one.

Example: with address = key MOD 10, keys 34 and 44 both hash to 4 (same last digit). The first one stored takes slot 4; the second collides.

There are two standard ways to resolve a collision:

1. Open addressing (next free slot). Keep everything in the main table: when the home slot is taken, search the following slots and store the record in the next free one (wrapping back to slot 0 if you reach the end). To read it back, re-hash to the home address and search forward until the matching key is found.

2. Overflow area. Reserve a separate region of the file for records that collide. A colliding record is written to this overflow area instead of the main table, and a pointer (or a search of the overflow area) is used to retrieve it later.

Whichever method is used, the exam answer must cover both halves: how the colliding record is stored and how it is found again afterwards.

Collision = same address for two different keys.
Open addressing: store in the next free slot; retrieve by re-hashing and searching forward.
Overflow area: store in a separate region; retrieve via pointer/search.
Always explain storage AND retrieval.

See the full worked example for file organisation and access →

Choosing the right organisation for a problem

Match the organisation to the access pattern the problem needs.

Exam questions almost always describe a scenario and ask you to choose an organisation and justify it. Work from the access pattern:

The problem needs…	Best organisation	Access method	Why
To append records as they arrive and later read the whole file	Serial	Sequential	Simple, fast appending; read all in order (e.g. a log)
To process every record once, in key order	Sequential	Sequential	Sorted by key field → efficient batch processing (e.g. payroll)
To fetch one record fast from a large file, by its key	Random	Direct	Address computed from the key → instant single-record access (e.g. ATM, online lookup)

A four-step routine for these questions:

Identify whether the problem needs the whole file or one record.
Decide if order matters (by a key field) or just arrival.
Pick the organisation from the table above.
Justify against the scenario — quote the requirement (e.g. 'must instantly fetch one record from millions') and link it to the organisation's strength.

Generic justifications score poorly; tie every reason back to the words in the question.

Append + read-all → serial (sequential access).
Process all in key order → sequential (sequential access).
Fast single-record lookup by key → random (direct access).
Always justify the choice against the scenario.

See the full worked example for file organisation and access →

Quick recap

Serial = arrival order (no key); sequential = key-field order; random = address from key via hashing.
Sequential access reads in order; direct access jumps to the record.
Serial + sequential files use sequential access; sequential + random files allow direct access.
A hashing algorithm maps a key to an address, e.g. address = key MOD numberOfSlots.
A collision is two different keys hashing to the same address.
Resolve collisions with open addressing (next free slot) or an overflow area — and explain retrieval.
Choose the organisation from the access pattern, and justify against the scenario.

Memorise this

Verbatim phrases and definitions Cambridge mark schemes credit.

Serial = order of arrival; sequential = order of a key field; random = hashed by record key.
Sequential access = read in order; direct access = go straight to the record.
Serial+sequential → sequential access; sequential+random → direct access.
Hashing algorithm: address = recordKey MOD numberOfSlots (MOD = remainder).
A 10-slot table uses addresses 0–9.
Collision = same address for two different keys.
Collision resolution: next free slot (open addressing) OR overflow area; always describe retrieval too.

How it’s examined

Section 13.2 is A-Level content assessed in Paper 3 (Advanced Theory, 1 h 30 min, 75 marks) — written, externally assessed, no calculator. Expect short structured questions: define/compare serial, sequential and random organisation; state which access methods each supports; apply a hashing algorithm (key MOD a small table size) to place several keys and handle a collision; and a 'choose and justify the organisation' scenario. Numbers are deliberately small so they can be done by hand — show your MOD working and always cover both storing and retrieving a colliding record.

Sources: Cambridge International AS & A Level Computer Science 9618 syllabus (2027-2029), section 13.2. Last reviewed 2026-06-21.

Take this whole topic with you

Step-by-step worked examples — File organisation and access

Step-by-step solutions to past-paper-style questions on file organisation and access, written exactly the way a tutor would explain them at the board.

Question type:

1Choosing an organisation method for an arriving stream of records
Getting startedWord problem• 9618 Paper 3 style• serial file, file organisation, AO2
Question
A logging system records each error message produced by a server. New error records simply need to be appended as they occur, and the whole log is later read through from start to finish to produce a report. State the most appropriate file organisation and justify your choice. (3 marks)
Step-by-step solution
1. Step 1
  Identify the access pattern. Records arrive over time and the whole file is later read in order — there is no need to jump to one particular record.
2. Step 2
  Match this to an organisation. A serial file stores records one after another in the order they arrive, with no key ordering. New records are simply added to the end.
3. Step 3
  Justify against the requirement. Appending is fast and simple (no need to find an insertion point), and reading the whole file from start to finish suits serial access perfectly.
Answer
Use a SERIAL file. Records are stored in arrival order and new ones are appended to the end, which is exactly what a log needs; the report is produced by reading the whole file sequentially.
Examiner tip
1 mark for naming the serial file, up to 2 for a justification tied to the scenario (appended in arrival order / whole file read in order). Saying 'sequential' here is a common slip — sequential means ordered by a key field, which this scenario does not require.
2Naming the access method for each file type
Getting startedIdentify & classify• 9618 Paper 3 style• sequential access, direct access, AO1
Question
For each of the following, state the method(s) of file access available: (a) a serial file; (b) a sequential file; (c) a random file. (3 marks)
Step-by-step solution
1. Step 1
  A serial file has no key order, so the only way to find a record is to read from the beginning until you reach it — that is sequential access.
2. Step 2
  A sequential file is ordered by a key field. You can read it in order (sequential access), but because it is ordered you can also jump in to find a record without reading every one — so it also supports direct access.
3. Step 3
  A random file places each record at an address computed from its key, so you can jump straight to the record — direct access.
Answer
(a) Serial file → sequential access only. (b) Sequential file → sequential access AND direct access. (c) Random file → direct access.
Examiner tip
The key discriminator examiners look for: sequential AND random files allow direct access, while serial AND sequential files can both be read using sequential access. A serial file cannot offer direct access because there is no key ordering or computed address.
3Serial vs sequential vs random for three requirements
Building confidenceWord problem• 9618 Paper 3 style• serial file, sequential file, random file, file organisation, AO2
Question
A company has three requirements: (a) a transaction log to which each sale is simply added as it happens; (b) a monthly payroll program that processes every employee record once, in employee-number order; (c) a help-desk system that must instantly fetch ONE customer's record from millions, given their customer ID. State and justify the most appropriate file organisation for each. (6 marks)
Step-by-step solution
1. Step 1
  Requirement (a): sales are only appended as they occur and there is no need to jump to one record or keep any key order. A serial file stores records in arrival order, so appending is simple and fast.
2. Step 2
  Requirement (b) processes ALL records, once each, in key order — order matters but single-record lookup does not. A sequential file is ordered by a key field (employee number), so reading it in order processes everyone efficiently — ideal for batch processing.
3. Step 3
  Requirement (c) needs ONE record fast, identified by a record key (customer ID), out of millions. Serial or sequential access would read far too many records.
4. Step 4
  A random file computes the record's address from its key using a hashing algorithm, giving near-instant direct access to that one record.
Answer
(a) SERIAL file — records appended in arrival order, which is all a log needs. (b) SEQUENTIAL file — ordered by the employee-number key field, ideal for processing every record once in order. (c) RANDOM file — the customer ID (record key) is hashed to an address for fast direct access to a single record.
Examiner tip
Marks split 2 + 2 + 2 (name + justification each). Discriminators: 'just append, no key' → serial; 'process all records in key order' → sequential; 'fast access to one record from many, by key' → random. Tie each justification to its scenario, not generic statements; the serial/sequential confusion (arrival order vs key-field order) is the usual fault line.
4Applying a hashing algorithm (key MOD table size)
Building confidenceDirect calculation• 9618 Paper 3 style• hashing algorithm, random file, AO2
Question
A random file uses a hash table with 10 slots, addresses 0 to 9. The hashing algorithm is address = recordKey MOD 10. Calculate the address used to store records with keys 47, 53 and 20. (3 marks)
Step-by-step solution
1. Step 1
  MOD gives the remainder after dividing by 10 — for a number, that is simply its last digit when dividing by 10.
  $\text{address} = \text{recordKey} \bmod 10$
2. Step 2
  Apply it to 47: 47 divided by 10 is 4 remainder 7.
  $47 \bmod 10 = 7$
3. Step 3
  Apply it to 53 and 20.
  $53 \bmod 10 = 3 \quad\text{and}\quad 20 \bmod 10 = 0$
Answer
Key 47 → address 7; key 53 → address 3; key 20 → address 0.
Examiner tip
1 mark per correct address. Always state the algorithm being used (key MOD 10). Note the addresses are 0–9, not 1–10 — a 10-slot table runs from slot 0.
5Handling a collision with open addressing
StretchMulti-step problem• 9618 Paper 3 style• hashing algorithm, collision, overflow, AO2
Question
A random file has 10 slots (0–9) and uses address = recordKey MOD 10. The records with keys 34 and 25 are already stored. A new record with key 44 now arrives. Show what happens and where 44 is finally stored, using open addressing (check the next free slot). (4 marks)
Step-by-step solution
1. Step 1
  Find the home addresses of the records already stored.
  $34 \bmod 10 = 4 \quad\text{and}\quad 25 \bmod 10 = 5$
2. Step 2
  So slot 4 holds key 34 and slot 5 holds key 25. Now hash the new key 44.
  $44 \bmod 10 = 4$
3. Step 3
  Key 44 hashes to address 4, but slot 4 is already occupied by key 34 — this is a COLLISION (two keys map to the same address).
4. Step 4
  Resolve with open addressing: check the next slot (5). Slot 5 is occupied (key 25), so move on to slot 6, which is free. Store key 44 in slot 6.
Answer
Key 44 hashes to address 4, which collides with key 34. Using open addressing the next free slot is searched: slot 5 is taken, so 44 is stored in slot 6.
Examiner tip
Marks for: computing 44 → 4 (1), identifying the collision (1), describing the next-free-slot search past slot 5 (1), final answer slot 6 (1). To read 44 back later, the system re-hashes to 4 and searches forward in the same way until it finds the matching key.
6Hashing with an overflow area
StretchMulti-step problem• 9618 Paper 3 style• hashing algorithm, collision, overflow area, AO2
Question
A random file uses address = recordKey MOD 7 for a main area of 7 slots (0–6). Records with keys 14 and 22 are already stored. A new record with key 21 arrives. (a) Show the collision. (b) Explain how an OVERFLOW AREA is used to resolve it, and how the record is later found. (4 marks)
Step-by-step solution
1. Step 1
  Find where the existing records live.
  $14 \bmod 7 = 0 \quad\text{and}\quad 22 \bmod 7 = 1$
2. Step 2
  Hash the new key 21.
  $21 \bmod 7 = 0$
3. Step 3
  Key 21 hashes to address 0, but slot 0 already holds key 14 — a collision. The main slot is full.
4. Step 4
  With an overflow-area strategy, the colliding record (21) is placed in a separate overflow area instead of the main table. A link/pointer from the home slot, or a search of the overflow area, lets the record be found later: the system hashes 21 → 0, sees it is not in slot 0, then looks in the overflow area for the matching key.
Answer
(a) 21 hashes to address 0 which is already occupied by key 14 — a collision. (b) Key 21 is written to the OVERFLOW AREA (a reserved region for records that cannot fit at their home address). To read it back, the system hashes 21 → 0, finds the wrong key there, then searches the overflow area for key 21.
Examiner tip
Distinguish the two resolution strategies cleanly: open addressing keeps everything in the main table by probing the next free slot, whereas an overflow area is a SEPARATE region holding records that collided. Both must explain how the record is retrieved later (re-hash, then search), not just how it is stored.

Model Answers — File organisation and access

High-scoring sample answers for file organisation and access on the Cambridge IGCSE paper, with examiner-style notes mapping each response to the mark scheme and assessment objectives.

Question 1
9618 Paper 3 style2 marks
Explain the difference between a serial file and a sequential file. (2 marks)
Model answer
In a serial file, records are stored one after another in the order in which they arrive (chronological order), with no ordering on any key.

In a sequential file, records are stored in order of a key field (for example ascending account number), so the records follow a meaningful sorted order rather than arrival order.
Why this scores
1 mark for serial = stored in order of arrival / no key order; 1 mark for sequential = stored in order of a key field. The phrase 'key field' is the high-value point — without it the second mark is usually lost.
Question 2
9618 Paper 3 style3 marks
Describe the difference between sequential access and direct access to a file. (3 marks)
Model answer
Sequential access means the records are read one after another from the start of the file until the required record (or the end) is reached; to reach the 100th record you must pass the 99 before it.

Direct access (also called random access) means a particular record can be read straight away without reading the records before it, because its position is known — for example its address has been computed from its key by a hashing algorithm.

Direct access is therefore much faster when only one specific record is needed, whereas sequential access suits reading or processing the whole file in order.
Why this scores
Marks for: sequential = read in order from the start (1); direct = go straight to the record without reading others (1); a sensible comparison of when each is suitable (1). Naming direct access as 'random access' is accepted.
Question 3
9618 Paper 3 style4 marks
A company stores customer records in a random file. (a) State what a hashing algorithm does. (b) Explain why a random file with hashing allows faster access to a single record than a serial file. (4 marks)
Model answer
(a) A hashing algorithm takes the record key (e.g. a customer number) and performs a calculation on it to produce a storage address in the file. For example, address = key MOD numberOfSlots. The same key always produces the same address, so the record can be placed and later found at that address.

(b) In a random file the address of a record is calculated directly from its key, so the system jumps straight to that record — direct access. In a serial file the records are in arrival order with no computed address, so to find one record the program must read through the file from the beginning (sequential access), which is much slower, especially when the file is large.
Why this scores
(a) 2 marks: takes the key / applies a calculation (1), produces a storage address (1). (b) 2 marks: random file computes the address → direct access (1); serial file must be searched from the start → slower (1). A worked formula such as 'key MOD 10' strengthens part (a).
Question 4
9618 Paper 3 style5 marks
A random file uses the hashing algorithm address = recordKey MOD 100. (a) Explain what is meant by a collision. (b) Describe two different methods that could be used to deal with a collision. (5 marks)
Model answer
(a) Collision: A collision occurs when the hashing algorithm produces the same address for two different record keys. For example, keys 150 and 250 both give 150 MOD 100 = 50, so both 'want' to be stored at address 50, but the slot can only hold one record.

(b) Method 1 — open addressing (next free slot): When the home address is already occupied, the algorithm searches the following addresses in turn and stores the record in the next free slot. To read the record back, it re-hashes to the home address and searches forward until the matching key is found.

Method 2 — overflow area: A separate overflow area is reserved for records that collide. When a record cannot be stored at its home address it is written to the overflow area, and a pointer or a search of the overflow area is used to retrieve it later.
Why this scores
(a) 2 marks: same address produced (1) for two different keys / a worked example (1). (b) up to 3 marks across two distinct methods — open addressing (next free slot) and an overflow area are the standard pair. Each method needs the store AND retrieve idea to gain full credit.
Question 5
9618 Paper 3 style6 marks
A hash table has 10 slots, numbered 0 to 9, all initially empty. The hashing algorithm is address = recordKey MOD 10. Collisions are resolved using open addressing (the next free slot, wrapping back to slot 0 if needed). The keys 23, 33, 43 and 30 are inserted in that order. (a) Show the slot in which each key is stored. (b) Explain why two of these keys end up away from their home address. (6 marks)
Model answer
(a) Apply address = key MOD 10 in order:

23 → 23 MOD 10 = 3. Slot 3 is free → store 23 in slot 3.

33 → 33 MOD 10 = 3. Slot 3 is taken (collision) → next free slot is slot 4 → store 33 in slot 4.

43 → 43 MOD 10 = 3. Slot 3 taken, slot 4 taken (collision) → next free is slot 5 → store 43 in slot 5.

30 → 30 MOD 10 = 0. Slot 0 is free → store 30 in slot 0.

Slot 0 1 2 3 4 5 6 7 8 9
Key 30 23 33 43

(b) The keys 33 and 43 all hash to the same home address (3) as 23, because 23, 33 and 43 share the same last digit, so 23 MOD 10 = 33 MOD 10 = 43 MOD 10 = 3. These are collisions. Open addressing therefore stores the later-arriving keys in the next free slots (4 and 5), away from their home address 3.
Why this scores
(a) up to 4 marks for the correct addresses with collision handling (allow follow-through if a mistake is carried consistently). (b) up to 2 marks: same home address because same last digit / 23≡33≡43 (mod 10) (1), so collisions push later keys to the next free slots (1). The key insight is that MOD 10 keys on the final digit.
Question 6
9618 Paper 3 style6 marks
A hospital must (i) keep an audit trail that simply appends every action as it happens, and (ii) instantly retrieve any single patient's record given their patient ID from a database of two million patients. For EACH requirement, state the most appropriate file organisation and access method, and justify your choices. (6 marks)
Model answer
(i) Audit trail — SERIAL file, sequential access. Actions only need to be appended in the order they occur, with no requirement to jump to a specific record. A serial file stores records in arrival order, so appending is simple and fast, and the trail is later read through in order using sequential access. There is no key to order by, so a sequential or random organisation would add needless complexity.

(ii) Patient retrieval — RANDOM file, direct access. A single record must be found instantly from two million, identified by the patient ID (the record key). A hashing algorithm computes the storage address from the patient ID, giving direct access so the record is fetched without searching the whole file. A serial or sequential file would be far too slow for fast single-record lookup at this scale.
Why this scores
Each requirement is worth up to 3 marks: name the organisation + access method (up to 2) and justify against the scenario (up to 1). The discriminators are 'append in order, read whole file' → serial/sequential access vs 'one record fast from millions, by key' → random/direct access. Generic justifications not tied to the scenario cap the marks.

Key Definitions and Keywords — File organisation and access

Definitions to memorise and the exact keywords mark schemes credit for file organisation and access answers — sharpened from recent examiner reports for the 2026 Cambridge IGCSE sitting.

Serial file
Examiner keyword
A file in which records are stored one after another in the order in which they arrive (chronological order), with no ordering based on any key.
Example
An error log that appends each new entry to the end of the file.
Sequential file
Examiner keyword
A file in which records are stored in order of a key field (for example ascending account number), so the records are kept in a meaningful sorted order.
Random (direct) file
Examiner keyword
A file in which the storage position of each record is computed from its record key by a hashing algorithm, allowing a record to be accessed directly without reading the others.
Key field
Examiner keyword
The field whose value is used to order the records in a sequential file (and to uniquely identify each record), such as an employee number.
Record key
Examiner keyword
The value that uniquely identifies a record and is fed into the hashing algorithm to calculate the record's storage address in a random file.
Sequential access
Examiner keyword
Accessing records one after another from the start of the file until the required record (or the end) is reached. Used for serial and sequential files.
Direct (random) access
Examiner keyword
Accessing a particular record straight away, without reading the records before it, because its position is known or can be computed. Available for sequential and random files.
Hashing algorithm
Examiner keyword
A calculation that takes a record key and produces a storage address in the file, e.g. address = key MOD numberOfSlots. The same key always produces the same address.
Example
address = recordKey MOD 10 maps key 47 to address 7.
Collision
Examiner keyword
The situation where the hashing algorithm produces the same address for two different record keys, so they cannot both be stored at that home address.
Overflow area
Examiner keyword
A separate region of the file reserved for records that collide and cannot be stored at their home address; a pointer or search is used to retrieve them later.
Open addressing (next free slot)
A collision-resolution method that keeps all records in the main table by searching forward from the home address and storing the record in the next free slot.
Home address
The address a hashing algorithm computes for a record key — the slot the record is placed in unless a collision forces it elsewhere.

Common Mistakes and Misconceptions — File organisation and access

The traps other students keep falling into on file organisation and access questions — taken from recent Cambridge IGCSE examiner reports and mark schemes — and how to avoid them.

✕Using 'serial' and 'sequential' interchangeably, or saying a sequential file is just stored 'in order'.
Examiner reports repeatedly note candidates confusing serial and sequential organisation.
Why it happens
Both words sound similar and both involve order, so students do not realise serial = arrival order while sequential = key-field order.
How to avoid it
Always state the distinguishing fact: serial = order of ARRIVAL (no key); sequential = order of a KEY FIELD. Mention the key field explicitly.
✕Claiming a serial file allows direct access to a record.
Why it happens
Students assume any file can jump to a record, forgetting a serial file has no key order and no computed address.
How to avoid it
Remember: serial and sequential files support sequential access; sequential and random files support direct access. A serial file offers sequential access only.
✕Numbering a table of N slots from 1 to N (or computing key MOD N but expecting addresses 1..N).
Why it happens
Students are used to counting from 1, but MOD N produces remainders 0 to N−1.
How to avoid it
A table of 10 slots uses addresses 0 to 9. key MOD 10 can only ever give 0–9, so start the table at slot 0.
✕Confusing MOD (remainder) with integer division, e.g. writing 47 MOD 10 = 4.
Why it happens
DIV and MOD are easily swapped; 47 DIV 10 = 4 but 47 MOD 10 = 7.
How to avoid it
MOD gives the REMAINDER after dividing. For MOD 10 it is simply the last digit: 47 MOD 10 = 7, 53 MOD 10 = 3.
✕Describing a collision as 'when the file is full' or 'when there is no space'.
Why it happens
Students associate the problem with running out of room rather than the precise hashing cause.
How to avoid it
Define a collision precisely: when the hashing algorithm produces the SAME address for two DIFFERENT keys. It can happen even in a mostly empty table.
✕Explaining how a collision is STORED (next free slot / overflow area) but never how the record is found again.
Why it happens
Students focus on writing the record and forget reading is the harder half.
How to avoid it
Always finish the answer with retrieval: re-hash the key to its home address, then search forward (open addressing) or search the overflow area until the matching key is found.

File organisation and access

Detailed Study Notes

At a glance

What you’ll learn

Quick recap

Memorise this

How it’s examined

Take this whole topic with you

1Choosing an organisation method for an arriving stream of records

2Naming the access method for each file type

3Serial vs sequential vs random for three requirements

4Applying a hashing algorithm (key MOD table size)

5Handling a collision with open addressing

6Hashing with an overflow area

Serial file

Sequential file

Random (direct) file

Key field

Record key

Sequential access

Direct (random) access

Hashing algorithm

Collision

Overflow area

Open addressing (next free slot)

Home address

✕Using 'serial' and 'sequential' interchangeably, or saying a sequential file is just stored 'in order'.

✕Claiming a serial file allows direct access to a record.

✕Numbering a table of N slots from 1 to N (or computing key MOD N but expecting addresses 1..N).

✕Confusing MOD (remainder) with integer division, e.g. writing 47 MOD 10 = 4.

✕Describing a collision as 'when the file is full' or 'when there is no space'.

✕Explaining how a collision is STORED (next free slot / overflow area) but never how the record is found again.