Table of Contents Link to heading

• Traditional and Physical Data Storage
• File-Based Digital Storage: An Improvement with Limits
  • Advantages
  • Disadvantages
• Where Databases Are Used

Traditional and Physical Data Storage Link to heading

Before digital systems, organisations managed data through physical records — card catalogues, filing cabinets, paper ledgers. This approach had fundamental structural constraints:

1. Physical capacity limits — storage is bounded by physical space
2. Single-user access — a borrowed book or checked-out file can only be used by one person at a time
3. Inflexible search — retrieval is limited to pre-built indexes (title, author, subject); cross-referencing requires manual effort
4. Slow retrieval — locating a record requires physically navigating the index system
5. Index maintenance overhead — every index must be kept synchronised with every other; a missed update in one creates inconsistency across all
6. Fragility to human error — a misplaced card, a misfiled document, or a deletion from one index but not another corrupts the entire system

Image Source: Britannica

File-Based Digital Storage: An Improvement with Limits Link to heading

The first wave of digitalisation moved data into flat files — spreadsheets, text files, and word processor documents. This eliminated physical space constraints and enabled faster search, but introduced a different set of problems.

Image Source: Orange Matter

Advantages Link to heading

• Low cost — requires no specialised infrastructure beyond existing file systems
• No additional software needed for basic storage and retrieval

Disadvantages Link to heading

1. Scale degrades performance — as file size grows, retrieval slows non-linearly; large flat files become impractical for querying
2. Concurrency and synchronisation — files are not designed for simultaneous multi-user access; concurrent edits produce version conflicts, overwrites, or data loss
3. Data consistency — duplicating data across multiple files (the only way to share it without a database) leads to inconsistencies when one copy is updated and others are not
4. Access control granularity — restricting access to specific fields (tax file numbers, passwords, financial records) within a flat file is difficult or impossible without application-level workarounds

These limitations are precisely what a Database Management System is designed to address.

Where Databases Are Used Link to heading

Database systems are the backbone of virtually every data-intensive application. The common thread is data that is highly valuable, large in volume, and accessed concurrently by multiple users or systems:

• Enterprise information

  • Sales: customer records, product catalogues, purchase history
  • Accounting: transactions, balances, assets, audit trails
  • Human resources: employee records, payroll, benefits

• Manufacturing: supply chain management, production tracking, warehouse inventory, order management

• Banking and finance

  • Core banking: accounts, loans, transaction ledgers
  • Credit cards: purchase records, billing cycles, fraud detection
  • Trading platforms: real-time market data, order books, portfolio holdings

• Universities: student enrolment, course registrations, grade records, academic history

• Airlines: reservations, scheduling, seat inventory — among the earliest adopters of distributed database systems

• Telecommunications: call detail records, data usage billing, prepaid balance management, network topology

• Web-based services

  • Social media: user graphs, posts, interactions, engagement metrics
  • E-commerce: product catalogues, order fulfilment, user behaviour for recommendation engines
  • Online advertising: click history, targeting profiles, impression tracking

• Document databases: research paper archives, patent repositories, news article collections

• Navigation systems: point-of-interest data, road network graphs, real-time routing

Adapted from Database System Concepts (2019) by Silberschatz, Korth, and Sudarshan. → PDF Version