week 3 - IMDBS

True

True or false: Before modern databases, every department ran its own software and maintained its own isolated data files. There was no single source of truth.

False

True of False: Users and applications CAN touch the Database directly. They only talk to the DBMS.

Database

The actual collection of data. It is passive. It’s just the organized pile of facts sitting on the hard drive.

• Analogy: The DB is like a massive, highly organized library building full of books.

Database Management System

The software that creates, manages, and interacts with the database. It is active. Examples: MySQL, Oracle, PostgreSQL, Microsoft SQL Server.

• Analogy: The DBMS is the Head Librarian. You don’t go into the library and start ripping books off shelves yourself. You ask the Librarian (DBMS) to find the book, check if it's available, and protect it from being stolen.

Record / Tuple

A single horizontal row. Represents one specific instance of an entity (e.g., the student Maria Santos).

Field / Attribute

A vertical slice. Represents a specific characteristic of the entity (e.g., First_Name).

Metadata / System Catalog

Data about data. The hidden rules defining the table structure (e.g., First_Name must be text, max 50 chars).

Uncontrolled Redundancy & Inconsistency

If an employee moves to a new house, HR updates their address in employees.txt. But Payroll doesn't know, so they mail the paycheck to the old house. The data is now inconsistent.

Data Isolation

It is incredibly difficult to write a program that pulls data from a .txt file, a .csv file, and an Access database at the same time to generate a cross-department report.

“all or nothing“

 In a file system, security is___________

Security Weaknesses

In a file system, security is "all or nothing. If a clerk needs to see an employee's name to print a name tag, they get access to the whole file—including the employee's salary and social security number.

Concurrency Issues

If two users try to update the exact same record in a text file at the exact same millisecond, the file corrupts or one update overwrites the other.

Controlled Redundancy

The DBMS stores the employee's address once. HR and Payroll just reference it..

Data Consistency

Update the address once, it updates for everyone instantly.

Granular Security

The DBMS can say, "Clerks can read the Name column, but only Managers can read the Salary column."

Concurrency Control

The DBMS uses "locks" so if User A is updating a record, User B has to wait in line. No overwritten data.

locks

The DBMS uses "_______" so if User A is updating a record, User B has to wait in line. No overwritten data.

Controlled Redundancy & Consistency

Data is stored once. Update an address, and it instantly updates for all departments.

Relational Databases (RDBMS)

• The Gold Standard: Data is stored in highly structured tables (relations) linked by common keys.

• Language: SQL (Structured Query Language).

• Characteristics: Strict rules, ACID compliance (ensures transactions are reliable), highly structured schema.

• Use Cases: Financial systems (Banks), ERPs, University Enrollment, Inventory systems.

• Examples: MySQL, PostgreSQL, Oracle, Microsoft SQL Server.

NoSQL Databases

• The Disruptor: "Not Only SQL." Born in the 2000s to handle massive, unstructured Web 2.0 data (social media, IoT) that didn't fit neatly into rows and columns.

• Characteristics: Flexible schemas, horizontal scalability (just add more cheap servers).

• Types:

  • Document (e.g., MongoDB): Stores data in JSON-like documents. Great for catalogs where products have wildly different attributes.

  • Key-Value (e.g., Redis): Extremely fast. Used for shopping carts or caching.

  • Graph (e.g., Neo4j): Stores relationships as first-class citizens. Perfect for social networks (who is friends with whom).

• Use Cases: Social media feeds, real-time big data, recommendation engines.

Document

What type of NoSQL Database does Stores data in JSON-like documents. Great for catalogs where products have wildly different attributes. (e.g., MongoDB)

Key-Value

What type of NoSQL Database is Extremely fast. Used for shopping carts or caching. (e.g., Redis)

Graph

What type of NoSQL Database does Stores relationships as first-class citizens. Perfect for social networks (who is friends with whom). (e.g., Neo4j)

MySQL

What language does Relational Databases use?

grandparents

Hierarchical and Network databases are considered the "_________" of databases from the 1960s.

family tree

The hierarchical database model looks like a strict _________ (one parent, many children).

Hierarchical

It looks like a strict family tree (one parent, many children)

Network

It allows multiple parents

Relational

Hierarchical and Network legacy databases were mostly replaced by _________ databases.

Gold Standard

On the "Mapping the Database Universe" diagram, Relational databases (RDBMS) are labeled as the "_________".

Web 2.0 Disruptor

NoSQL databases are labeled as the "_________".

Demployment Model

The term "Cloud" refers to a _________ model, not a structural model.

NoSQL

Cloud databases are simply Relational or _________ systems that are hosted on vendor hardware.

Relational

Services like AWS RDS or Azure SQL are actually just _________ databases hosted on Amazon/Microsoft servers.

vendor

A major benefit of using Cloud databases is that the _________ handles the hardware, power, cooling, and backups, while you just pay a monthly fee and start writing SQL.