The File-Based Storage Nightmare

Before databases (DBMS) ruled the world, we had File-Based Storage Systems. Essentially, data was stored in folders and text files. While simple, it falls apart quickly at scale.

What is a File-Based System?

Imagine storing college student records in a folder structure:

Folder: Computer Science -> File: Ram.txt
Folder: Electrical Engineering -> File: Shyam.txt

The 4 Major Challenges

1. Data Redundancy

The same data is stored in multiple places.

Example: “Ram” takes both CS and Math classes. His address is stored in CS/Ram.txt AND Math/Ram.txt.
Result: Wasted storage.

2. Data Inconsistency

Since data is duplicated, updates might fail on one copy.

Scenario: Ram moves to a new house. You update CS/Ram.txt but forget Math/Ram.txt.
Result: The system has two different addresses for the same person. Which one is true?

  %%{init: {'theme': 'dark'}}%%
graph TD
    User["User"] --> Update["Update Address"]
    Update --> File1["CS/Ram.txt (Updated)"]
    Update -.->|Forgot to Update| File2["Math/Ram.txt (Old Address)"]
    style File1 fill:#99ff99,stroke:#333
    style File2 fill:#ff9999,stroke:#333

3. Security Issues

It’s hard to control granular access.

Scenario: You want a professor to see Ram’s grades but not his home address.
Problem: In a file system, if you give read access to the file, they see everything. You can’t easily hide specific “columns”.

4. Slow Data Access

No indexing.

Scenario: “Find all students with GPA > 3.5”.
Problem: You have to open every single file and check. This is incredibly slow compared to a SQL query.

Conclusion

File-based systems are fine for simple, personal data. But for any application with multiple users and relationships, you need a Relational Database Management System (RDBMS) to handle Consistency, Security, and Speed.