Advanced Data Representation and File Organisation: Question 2

Worked solution

Part (a): JobLog

JobLog has two defining features: new records are only ever appended to the end, and the whole file is read straight through from start to finish, never searched for one particular job. There is no key field being used to keep the file in any particular order.

This is exactly what serial file organisation provides: records are stored one after another in the order they are written, with no attempt to order them by a key. The matching access method. Reading every record in turn from the beginning, is called sequential access.

[2 marks]: [1] for stating serial file organisation, [1] for a justification that connects “always appended to the end” and/or “always read straight through” to that choice.

Part (b): CustomerMaster

CustomerMaster has records held in order of the key field CustomerID, and needs to support two different needs: a full pass through every record in key order (the monthly update), and an immediate single-record lookup (the receptionist’s phone call).

Sequential file organisation. Where records are physically stored in ascending order of the key field, is the appropriate choice, because unlike a random file it preserves a meaningful overall order, and unlike a serial file that order is based on the key rather than arrival time.

Crucially, a sequential file supports both of the file access methods the garage needs:

• Sequential access: reading through the file from the first record to the last, in CustomerID order. Used for the monthly loyalty-points update, which must touch every record.
• Direct access: locating one specific customer’s record immediately, for example using an index built on CustomerID that maps a key value straight to its position in the file, used for the receptionist’s immediate lookup, without reading through every preceding record.

[3 marks]: [1] for stating sequential file organisation, [1] for identifying that it supports sequential access (linking this to the monthly update), [1] for identifying that it also supports direct access (linking this to the immediate phone lookup).

Part (c): PartsStock

PartsStock needs a single record fetched immediately, by its key (PartCode), with no other records read at all. Neither serial file organisation (which only supports reading from the start) nor sequential file organisation (which, even with direct access, still relies on an ordered structure or index built up over the whole file) is as directly suited to this as:

Random file organisation, which supports direct access exclusively: the storage location of a record is calculated straight from its key field using a hashing algorithm. The hashing algorithm takes PartCode as input and produces a value that identifies exactly where that record is stored (e.g. which record or block number in the file). The till system can then SEEK straight to that location and read the one record it needs, without processing any other record in the file.

[3 marks]: [1] for stating random file organisation, [1] for identifying that this gives direct access, [1] for explaining that a hashing algorithm calculates the storage location directly from the key.

Final answers

• (a) Serial file organisation. Records are appended in arrival order and read straight through.
• (b) Sequential file organisation (ordered by CustomerID), supports both sequential access (the monthly update) and direct access (the receptionist’s immediate lookup).
• (c) Random file organisation, a hashing algorithm calculates a record’s storage location directly from its key (PartCode), giving immediate direct access.