Unit-1

Database System
Architecture
 Outline
Looping
• Introduction of DBMS
• Applications of DBMS
• Advantages of DBMS
• Three levels ANSI SPARC database system
• Data Abstraction in DBMS
• Mappings and data independence
• Database users and DBA
• Database system architecture
Evaluation policy
 Theory Marks (100)
 End Semester Examination: 50 Marks
 Continuous Evaluation : 50 Marks

Quiz 1 10
Mid-sem exam 20
Quiz 2 10
Class participation 5
Attendance 5

 Practical Marks (50)

 Practical Assessment : 25 Marks
 Viva : 25 Marks

Unit 1 – Database System Architecture 3

Section - 1
What is Database Management System (DBMS)?
 Data - Fact that can be recorded or stored
 e.g. Person Name, Age, Gender and Weight etc.
 Database - Collection of logically related data
 e.g. Books Database in Library, Student Database in University etc.
 Management - Manipulation, Searching and Security of data
 e.g. Viewing result in GTU website, Searching exam papers in GTU website etc.
 System - Programs or tools used to manage database
 e.g. SQL Server Studio Express, Oracle etc.
 DBMS - A Database Management System is a software for creating and managing databases.
 Database Management System (DBMS) is a software designed to define, manipulate, retrieve
and manage data in a database.
 e.g. MS SQL Server, Oracle, My SQL, SQLite, MongoDB etc.

Unit 1 – Database System Architecture 5

Section - 2
Applications of DBMS
 DBMS is a computerized record-keeping system.
 DBMS is required where ever data need to be stored.
 E-Commerce (Flikart, Amazon, Shopclues, eBay etc...)
 Online Television Streaming (Hotstar, Amazon Prime etc...)
 Social Media (WhatsApp, Facebook, Twitter, LinkedIn etc...)
 Banking & Insurance
 Airline & Railway
 Universities and Colleges/Schools
 Library Management System
 Human Resource Department
 Hospitals and Medical Stores
 Government Organizations

Exercise Write down any five applications of DBMS other than above.

Unit 1 – Database System Architecture 7

Section - 3
Reduce data redundancy (duplication)

Computer Civil

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234 DBMS Prof. ABC India 1234 DBMS

Database management system

can remove such data Same data is stored at
redundancy by storing data four different places.
centrally.

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234 DBMS Prof. ABC India 1234 DBMS

Electrical Mechanical

Unit 1 – Database System Architecture 9

Remove data inconsistency

Computer Civil

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234
6789 DBMS Prof. ABC India 1234 DBMS

Same data having

Database management different state (values)
system can keep data in
consistent state.
Mobile no is changed

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234
6789 DBMS Prof. ABC India 1234 DBMS

Electrical Mechanical

Unit 1 – Database System Architecture 10

Data isolation
 Data are scattered in various files.
File - 1
 Files may be in different formats.
Emp_Name Address Mobile Subject
 Difficult to retrieve the appropriate data. Prof. ABC India 1234 DBMS

DBMS allow us to access (retrieve) appropriate data easily. File - 2

Emp_Name Post Salary Load
Prof. ABC Prof 90,000 15
Data isolation is a property that determines when and how
changes made by one operation become visible to other
concurrent users and systems. File - 3
This issue occurs in a concurrency situation. Emp_Name Teaching Knowledge Rating
Prof. ABC Good Excellent 9

Unit 1 – Database System Architecture 11

Guaranteed atomicity
 Atomicity: Either transaction execute 0% or 100%.

Sum of both account

before transfer is 3000

Person A Person B
Account A Transfer 500 Account B
Bal : 2000 Bal : 1000
Step 1 : Debit 500 from Account A
Step 2 : Credit 500 into Account B
Transaction Sum of both
is failed Sum of both account account is 2500
after transfer is 3000 so inconsistent

Unit 1 – Database System Architecture 12

Allow to implement integrity constraints

Emp_Name Address Mobile_No Subject

Prof. ABC India 98XXXXXXXX DBMS

Should contain exact 10 digits

Student_Name Branch Backlog SPI

Nirav Shah Civil 0 8.5

Should be between 0 to 10

DBMS allows us to implement such business rules in our database..

Unit 1 – Database System Architecture 13

Sharing of data among multiple users

Want to access
Computer Civil

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234 DBMS Prof. ABC India 1234 DBMS

Want to access Database management system

allows more than one user to
access same data simultaneously.

Emp_Name Address Mobile Subject Emp_Name Address Mobile Subject

Prof. ABC India 1234 DBMS Prof. ABC India 1234 DBMS

Electrical Mechanical

Unit 1 – Database System Architecture 14

Restricting unauthorized access to data

File - 1
Emp_Name Address Mobile Subject
Prof. ABC India 1234 DBMS
Wants to
access Faculty
File - 2 of other
Emp_Name Post Salary Load college
Prof. ABC Prof 90,000 15
Wants to
access Adani Institute
File - 3 Faculty
Emp_Name Teaching Knowledge Rating
Prof. ABC Good Excellent 9

DBMS prevents unauthorized user to access data.

Unit 1 – Database System Architecture 15

Providing backup and recovery services

Provides facilities to backup and restore the database in case of failure.

Unit 1 – Database System Architecture 16

Advantages of DBMS (Summary)
 Reduce data redundancy (duplication)
 Avoids unnecessary duplication of data by storing data centrally.
 Remove data inconsistency
 By eliminating redundancy, data inconsistency can be removed.
 Data isolation
 A user can easily retrieve proper data as per his/her requirement.
 Guaranteed atomicity
 Either transaction executes 0% or 100%.

Unit 1 – Database System Architecture 17

Advantages of DBMS (Summary)
 Allow implementing integrity constraints
 Business rules can be implemented such as do not allow to store amount less than Rs. 0 in balance.
 Sharing of data among multiple users
 More than one users can access same data at the same time.
 Restricting unauthorized access to data
 A user can only access data which is authorized to him/her.
 Providing backup and recovery services
 Can take a regular auto or manual backup and use it to restore the database if it corrupts.

Unit 1 – Database System Architecture 18

Section - 4
Basic terms
 Data
 Data is raw, unorganized facts that need to be processed.
 Example: Marks of students
 Student_1 = 50/100, Student_2 = 25/100.
 Information
 When data is processed, organized, structured or presented in a given context so as to make it useful, it is
called information.
 Example: Result of students (Pass or Fail)
 Student_1 = Pass, Student_2 = Fail.

Unit 1 – Database System Architecture 20

Basic terms (cont…)
 Metadata
 Metadata is data about data.
 Data such as table name, column name, data type, authorized user and user access privileges for any table is
called metadata for that table.

Faculty
Emp_Name Address Mobile_No Subject
Prof. ABC India 98XXXXXXXX DBMS

 Metadata of above table is:

 Table name such as Faculty
 Column name such as Emp_Name, Address, Mobile_No, Subject
 Datatype such as Varchar, Decimal
 Access privileges such as Read, Write (Update)

Unit 1 – Database System Architecture 21

Basic terms (cont…)
 Data dictionary
 A data dictionary is an information repository which contains metadata.
• Table Name – Faculty
• Column Name – EmpName, Address, Mob, Subject, Salary
• Datatype – Varchar, Decimal
• Access Privileges – Read, Write (Update)

 Data warehouse
 A data warehouse is an information repository which stores data.
Faculty
Emp_Name Address Mobile_No Subject
Prof. ABC India 98XXXXXXXX DBMS
Prof. XYZ USA 01XXXXXXXX OS

Exercise Why data dictionary and data warehouse are stored in the different places?

Unit 1 – Database System Architecture 22

Basic terms (cont…)
 Field
 A field is a character or group of characters that have a specific meaning.
 E.g, the value of Emp_Name, Address, Mobile_No etc are all fields of Faculty table.
Faculty
Emp_Name Address Mobile_No Subject Fields
Prof. ABC India 98XXXXXXXX DBMS
Prof. ABC India 98XXXXXXXX
Prof. XYZ USA 01XXXXXXXX OS

 Record / Tuple
 A record is a collection of logically related fields.
 E.g, the collection of fields (Emp_Name, Address, Mobile_No, Subject) forms a record for the Faculty.

Prof. ABC India 98XXXXXXXX DBMS

Record / Tuple
Prof. XYZ USA 01XXXXXXXX OS

Unit 1 – Database System Architecture 23

Section - 5
3 Levels ANSI SPARC Database System

User 1 User 2 User 3

How data are viewed View
View 1 View 2 View 3
by each users? Level

What data are stored and Conceptual Logical

What relationships exist? Level Level

How the data are actually Internal Physical

stored on storage devices? Level Level

Database

Unit 1 – Database System Architecture 25

3 Levels ANSI SPARC Database System
 Internal level (Physical level)
 It describes how a data is stored on the storage device.
 Deals with physical storage of data.
 Structure of records on disk - files, pages, blocks and indexes and ordering of records
 Internal view is described by the internal schema.
 Conceptual level (Logical level)
 What data are stored and what relationships exist among those data?
 It hides low level complexities of physical storage.
 For Example, STUDENT database may contain STUDENT and COURSE tables which will be visible to users but
users are unaware about their storage.
 Database administrator works at this level to determine what data to keep in the database.
 External level (View level)
 It describes only part of the entire database that an end user concern or how data are viewed by each user.
 Different user needs different views of the database, so there can be many views in a view level abstraction
of the database. Used by end users and application programmers.
 End users need to access only part of the database rather than the entire database.
Unit 1 – Database System Architecture 26
3 Levels ANSI SPARC Database System: Example

We are storing student information in a student table.

User 1 User 2 User 3
User just interact with system with the help of GUI. View
View 1 View 2 View 3
Users are not aware of how and what the data is stored. Level

Records can be described as fields and attributes along

with their data types, their relationship among each other Conceptual
can be logically implemented. Logical
Level Level
Programmers generally work at this level.

Records can be described as blocks of storage (bytes,

Internal Physical
gigabytes, terabytes etc.) in memory.
Level Level
These details are often hidden from the programmers.

Database

Unit 1 – Database System Architecture 27

Data Abstraction in DBMS
 Database systems are made-up of complex data structures.
 To ease the user interaction with database, the developers hide internal irrelevant details from
users.
 This process of hiding irrelevant details from user is called data abstraction.

Unit 1 – Database System Architecture 28

Mapping and Data Independence
Want to access some data

User 1 User 2 User 3

View
View 1 View 2 View 3
Level

Request
Process of transforming requests and results between the
three levels is called mapping. Conceptual Logical
Level Level

Ability to modify a schema definition in one level without

Internal Physical
affecting a schema definition in the next higher level.
Level Level

Result
Database

Unit 1 – Database System Architecture 29

Types of Data Independence
 Physical Data Independence
 Physical Data Independence is the ability to modify the physical schema without requiring any change in
logical (conceptual) schema and application programs.
 Modifications at the internal levels are occasionally necessary to improve performance.
 Possible modifications at internal levels are changes in file structures, compression techniques, hashing
algorithms, storage devices, etc.
 Logical Data Independence
 Logical data independence is the ability to modify the conceptual schema without requiring any change in
application programs.
 Modification at the logical levels is necessary whenever the logical structure of the database is changed.
 Application programs are heavily dependent on logical structures of the data they access. So any change in
logical structure also requires programs to change.

Unit 1 – Database System Architecture 30

Section - 6
Types of Database Users
 Naive Users (End Users)
 Unsophisticated users who have zero knowledge of database system
 End user interacts to database via sophisticated software or tools
 e.g. Clerk in bank
 Application Programmers
 Programmers who write software using tools such as Java, .Net, PHP etc…
 e.g. Software developers
 Sophisticated Users
 Interact with database system without using an application program
 Use query tools like SQL
 e.g. Analyst
 Specialized Users (DBA)
 User write specialized database applications program
 Use administration tools
 e.g. Database Administrator

Unit 1 – Database System Architecture 32

Section - 7
Role of DBA
 Schema Definition
 DBA defines the logical schema of the database.
 Storage Structure and Access Method Definition
 DBA decides how the data is to be represented in the database & how to access it.
 Defining Security and Integrity Constraints
 DBA decides on various security and integrity constraints.
 Granting of Authorization for Data Access
 DBA determines which user needs access to which part of the database.

Unit 1 – Database System Architecture 34

Role of DBA
 Assisting Application Programmer
 DBA provides assistance to application programmers to develop application programs.
 Monitoring Performance
 DBA ensures that better performance is maintained by making a change in the physical or logical schema if
required.
 Backup and Recovery
 DBA backing up the database on some storage devices such as DVD, CD or magnetic tape or remote servers
and recover the system in case of failures, such as flood or virus attack from this backup.

Unit 1 – Database System Architecture 35

Section - 8
Database System Architecture
Naive Application Sophisticated Database
user programmer user administrator
uses write uses uses

Application Application Query Administration

interfaces program tool tool

Translates
Interprets DML
DDL
Compiler DML DDL statements intoa
and linker queries interpreter statements
Deals
Executes low into
with level
low
set oflevel
tables
Application execution of
instructions
instructions DDL
that
program DML compiler containing
and
generated DML
by DML
the query
metadata
object code and organizer statements
compiler.
Query evaluation evaluation engine
engine Query processor understands

Buffer File Authorization and Transaction

manager manager integrity manager manager

Manages allocation
Storage manager
Fetches data from
of space
disk on disk
storage to Preserves atomicity
Provides interface
storage
memory for being Checks the authority
and controls
between low-level
used faster Indices Data dictionary of users to access
To provide Disk storage concurrency
data stored and
data and integrity
access to data items Data Statistical data application program
To store user data To To store
store statistical
metadata constraints
or queries
information about the data

Unit 1 – Database System Architecture 37

Question Bank
1. List and explain the advantages of DBMS over file based system. OR Explain disadvantages of
files based system.
2. Draw and explain 3 level architecture of DBMS.
3. List and explain different categories/types of database users.
4. List and explain different tasks/roles/functions/duties of DBA (Database Administrator).
5. Explain DBMS architecture with block diagram. OR Explain Database System architecture with
block diagram.

Unit 1 – Database System Architecture 38

Thank
You