0.1.2 · What a DBMS is and what it does; its users (end users, application programmers, DBA)

0 · Recall Warm-up

Answer from memory first. Answers are in the Answer Vault at the bottom.

What is the difference between redundancy and inconsistency? from Lesson 0.1.1
A transfer subtracts money from X, then crashes before adding it to Y. Which problem? from Lesson 0.1.1
Is "average CGPA of EE students = 7.8" data or information? from Lesson 0.1.1

1 · The Hook

Think of a railway reservation system. Lakhs of people search trains, book seats and cancel tickets at the same time. Nobody gets the same seat twice, a crash in the middle of a booking never takes money without giving a ticket, and a booking clerk cannot see the railway's salary records.

None of this is done by the booking app itself. It is done by one piece of software that sits between every app and the stored data: a database management system. This lesson says what that software does and who uses it.

2 · Predict First

A college moves from separate files to one database managed by a DBMS. Of the eight problems from the last lesson (RIDIIACS), which ones can a DBMS fix completely, and is there anything it makes worse?

Make a guess in your head. The answer opens Section 3.

3 · Core Concept

Predict First — the answer

A DBMS controls redundancy (each fact stored once, or extra copies kept in step on purpose), lets new questions be asked without new programs, keeps rules in one place, makes tasks all-or-nothing, manages concurrent users and enforces access rights. So it addresses all eight.

It does make some things worse: a DBMS costs money and expertise to set up, and its general-purpose machinery adds overhead, so a tiny single-user program can be faster with a plain file.

Definitions

Database (DB): a collection of related data.
Database management system (DBMS): software that lets users define, build, query, update and share a database, while protecting it.
Database system: the database + the DBMS (+ the application programs that use them).

What the DBMS does

Job	Meaning	Fixes
Definition	Declare what data exists: record types, their fields, their types, their rules.	Integrity, isolation
Querying	Answer questions on demand in a high-level language.	Difficult access
Updating	Insert, delete and change data.	Redundancy (data stored once)
Concurrency control	Let many users work at once without losing changes.	Concurrent-access anomalies
Recovery	After a crash, undo half-finished tasks and keep finished ones.	Atomicity
Security	Give each user only the access they are allowed.	Security
Integrity checks	Reject data that breaks declared rules (CGPA 0–10).	Integrity

The catalog: a self-describing database

A DBMS stores not only the data but also a full description of it: which record types exist, their fields, types and rules. This description is called metadata ("data about data") and it is kept in the catalog (also called the data dictionary). Because the description lives inside the database, the DBMS is self-describing: any program can ask the DBMS what is stored. In a file system, the file format is buried inside each program.

Two kinds of language

A data definition language (DDL) declares the structure: "a student record has a roll number, a name and a CGPA".
A data manipulation language (DML) reads and changes the data: "give me all students with CGPA above 8".

(SQL, taught in Phase 3, is both a DDL and a DML.)

Who uses a database

Role	What they do
Database administrator (DBA)	In charge of the whole database: grants and revokes access, decides storage structures, monitors performance, plans backup and recovery.
Database designers	Decide what data to store and how to structure it, by talking to future users.
End users	Use the data. Naive (parametric) users press buttons in ready-made apps (a booking clerk). Casual users write an occasional query (a manager). Sophisticated users write complex queries themselves (an analyst). Stand-alone users keep a personal database (a tax app).
Application programmers	Write the programs that naive users run, using the DML inside a programming language.

When not to use a DBMS

Cost (software, hardware, training) and overhead (security, concurrency and recovery machinery that a tiny single-user application does not need). For a small, fixed, single-user application, plain files can be the right choice.

4 · Visual

Every user and program reaches the data only through the DBMS; the DBMS also keeps the description of the data (the catalog).

5 · Worked Examples

Example 1 Easy

Who would do each task: (a) grant a new clerk read-only access; (b) book a ticket through a website; (c) write the code behind the "Book" button?

Solution

(a) the DBA; (b) a naive (parametric) end user; (c) an application programmer.

Example 2 Medium

Why is a database called self-describing? What would a file-processing system need to do to answer "which fields does the hostel record have?"

Solution

Because the DBMS stores the metadata (record types, fields, types, rules) in its catalog, inside the database itself. In a file system, the record layout is hard-coded inside each program; you would have to read the program's source code to find the fields.

Example 3 GATE

Match each DBMS job with the file-processing problem it solves: (1) recovery, (2) concurrency control, (3) authorisation checks, (4) declared rules checked on every update. Problems: P = security, Q = atomicity, R = integrity, S = concurrent-access anomaly.

Solution

1–Q (recovery undoes half-finished tasks), 2–S, 3–P, 4–R.

6 · GATE Traps

Database ≠ DBMS. The database is the data; the DBMS is the software. "Database system" means both together.
Metadata is stored in the catalog, inside the database. It is not stored in the application programs.
A DBMS controls redundancy; it does not promise zero copies. Some copies are kept on purpose (for speed or backup) and the DBMS keeps them in step.
Naive users do not write queries; they use apps. Casual and sophisticated users write queries.
DBA is a role, not a program.

7 · Memory Hook

"DDL draws the map, DML drives on it." The catalog is the map's legend, kept inside the atlas itself, so the atlas describes itself.

8 · Connect the Dots

Lesson 0.1.1 listed eight problems; this lesson showed the DBMS solving each one and keeping a description of the data in its catalog. Next: that description is called a schema, and the DBMS keeps it at three levels so that changing how data is stored does not break the programs that use it.

9 · Self-Test

GATE format. Try every question before opening the vault.

Q1 · MCQ · 1 mark(s) — Definitions

Metadata in a DBMS is stored in the:

(A) application programs
(B) operating system's file table
(C) log of user queries
(D) catalog

Q2 · MCQ · 1 mark(s) — Users

A railway booking clerk who uses only the booking screens is a:

(A) application programmer
(B) sophisticated user
(C) naive (parametric) end user
(D) DBA

Q3 · MSQ · 2 mark(s) — DBMS jobs

Which are jobs of a DBMS?

(A) Checking that a CGPA entered is between 0 and 10, if that rule was declared
(B) Designing the college's admission policy
(C) Letting two users update data at the same time safely
(D) Undoing a half-finished task after a crash

Q4 · MCQ · 1 mark(s) — Languages

"Every student record has a roll number, a name and a CGPA." This statement belongs to the:

(A) DDL
(B) DML
(C) recovery log
(D) catalog query

Q5 · MCQ · 1 mark(s) — Self-describing

A database is called self-describing because:

(A) users must describe their data before every query
(B) the data has no fixed format
(C) every record contains its own field names
(D) the DBMS stores the definition of the data (metadata) in the database's catalog

10 · Rule Card

Database = related data; DBMS = software to define, build, query, update and share it; database system = both.
DBMS jobs: definition, querying, updating, concurrency control, recovery, security, integrity checks.
Metadata (data about data) lives in the catalog: the database is self-describing.
DDL defines structure; DML reads and changes data.
Users: DBA (access, storage, performance, backup), designers, end users (naive, casual, sophisticated, stand-alone), application programmers.
Costs of a DBMS: money, training, overhead.

11 · Concept Ledger

Unlocked in this lesson. From here on, these terms may be used.

DBMSdatabase systemmetadatacatalog (data dictionary)self-describingDDLDMLDBAdatabase designernaive / casual / sophisticated / stand-alone end userapplication programmerconcurrency controlrecovery

12 · Answer Vault

Reveal answers when you’re ready

Recall Warm-up

Redundancy: the same fact stored more than once. Inconsistency: those copies disagree.
An atomicity problem (the task was not all-or-nothing).
Information: it is computed from many raw values.