File organisation and access — common mistakes

A common exam question asks you to justify the choice of file organisation for a given scenario. Always link your choice back to the application's requirements. Does it need fast access to single records (random)? Or does it process all records in a batch (sequential)?

Be precise with your terminology. 'Serial' refers to unsorted chronological order. 'Sequential' refers to being sorted by a key. 'Direct' access is not the same as 'random' organisation, but it is enabled by it. Use these terms correctly to gain full marks.

The key difference is sorting. In serial organisation, records are simply added to the end in the order they arrive (unsorted). In sequential organisation, records are stored in order based on a key field (sorted). This means a sequential file is always sorted, while a serial file is not.

True direct access requires a mechanism to translate a record's key directly into a physical disk address. Sequential files don't have this. Even if you know you want the 50th record and all records are the same size, the system still has to start at the beginning and skip the first 49 records to get there. This is still a form of sequential access, not a direct jump to a calculated address.

Random organisation has its own trade-offs. It can be complex to implement, hashing algorithms can cause collisions which slow down the system, and it often leads to wasted disk space ('sparse files'). Furthermore, it's very inefficient for tasks that require processing records in order, like generating a report for all customers. The best method always depends on the specific application.

Yes, for the purposes of the A-Level syllabus, the terms 'Random File Organisation' and 'Direct File Organisation' are used interchangeably. They both refer to a system where a record's location can be calculated from its key, allowing for direct access.