Describe non-database methods of organising data, including flat files, hypermedia and free text retrieval, and the storage and retrieval methods used by information systems

What this dot point is asking

NESA wants you to describe the ways an information system organises data for storage and retrieval, and to compare them. You need to know flat file systems, the database approach, hypermedia and free text retrieval, and the physical and logical methods used to store data and bring it back efficiently.

The answer

Why organisation matters

Two of the seven information processes are organising and storing and retrieving. How data is organised decides how easily it can be searched, updated and kept consistent. The right structure depends on the kind of data and how it will be used, so the syllabus asks you to compare several approaches rather than treat the relational database as the only option.

Flat file systems

A flat file stores all data in a single table or file, with every record holding every field. A spreadsheet of customers and their orders is a flat file. It is simple to create and read, and fine for small, single-purpose data. Its weakness is redundancy: a customer who places ten orders has their name and address repeated ten times. That wastes space and creates update anomalies, where changing the address in one row but not the others leaves the data inconsistent.

The database approach

A database stores data in multiple linked tables, with each fact recorded once and connected to others through keys. This removes the redundancy of flat files, enforces consistency, and supports complex queries across related data. The cost is greater design effort and a database management system to maintain the structure. The database approach also separates the data from the programs that use it, so many applications can share one well-organised store.

Hypermedia

Hypermedia organises information as nodes of text, images, audio and video joined by navigable links. The World Wide Web is the largest example: pages connect to other pages through hyperlinks, and the user navigates rather than queries. Hypermedia suits browsing and exploration where the path through the information is not fixed in advance, but it is poor for the precise, structured retrieval a database gives.

Free text retrieval

Free text retrieval searches collections of unstructured documents, such as articles or emails, for records containing particular words or phrases. A search engine is the familiar example. The system builds an index of the words in every document so it can find matches quickly. Free text retrieval handles data that has no fixed fields, but it returns documents rather than precise data values, and can return irrelevant matches.

Storage and retrieval methods

Once organised, data must be stored on physical media and retrieved efficiently.

• Storage media include magnetic (hard disk, tape), optical (DVD, Blu-ray), solid state (SSD, flash) and cloud storage. Each trades off capacity, speed, cost and durability. Tape is cheap and high capacity but slow, suited to backup; solid state is fast but costlier per gigabyte.
• Online storage is immediately accessible by the system; offline storage (a tape in a vault) must be loaded before use. Near-line storage sits between the two.
• Indexes speed retrieval by acting like a book's index: rather than scanning every record, the system looks up the index to jump straight to the data. The trade-off is that indexes take space and must be updated when data changes.
• File organisation methods include sequential (records in order, read start to finish), and direct or random access (jump straight to a record by its key or address). Direct access is far faster for finding one record.

Backup and archiving

Responsible storage includes backup (a second copy for recovery after loss) and archiving (moving old but valuable data to cheaper long-term storage). A backup strategy specifies what is copied, how often, and where the copies are kept, ideally off site.