Comparison of RDBMS with File Based System, Hierarchical, Network and Relational Database Models

This article compares RDBMS with File Based System, Hierarchical, Network and Relational Database Models. It also covers Normalization and ER Diagram for AutoSellers scenario.

Contents
TASK 1.................................................................................................................................................2
Question 1.........................................................................................................................................2
Question 2.........................................................................................................................................3
Question 3.........................................................................................................................................4
TASK 2.................................................................................................................................................5
Question 1.........................................................................................................................................5
Question 2.........................................................................................................................................5
TASK 3.................................................................................................................................................6
Question 1.........................................................................................................................................6
Question 2.........................................................................................................................................7
Question 3.........................................................................................................................................8
TASK 4.................................................................................................................................................9
Question 1.........................................................................................................................................9
Question 2.......................................................................................................................................10
Bibliography.......................................................................................................................................11

TASK 1
Question 1
In attempting to compare RDBMS (relational database management systems) with the file based
system on their merits and demerits we should start with a brief description of both before we
compare them.
RDBMS: Relational Database management system is a software that allows for easy manipulation,
storage and retrieval of data. In an rdbms system, entities are represented as tables and records are
represented as rows. It (RDBMS) is accessed and manipulated using SQL (Structured Query
Language).
File System: The file system handles the way data is written to and read from the hard disk of the
computer. It is preinstalled with the operating system of the computer and different OS vendors
have their own filing system. Some examples are New Technology File System (NTSF) for
Windows, and Extended File System (Ext) for Linux.
The tables below enumerates the merits and demerits of each systems.
File System
Advantages Disadvantages
Data is more secure Data storage and retrieval is done manually.
Appropriate for small scale organisational needs Encourages data inconsistency and redundancy
Easy to handle Recovery of lost data virtually impossible
Table 1.1 File System
RDBMS
Advantages Disadvantages
Data storage and manipulation is easy and
efficient using SQL
Handling is complex
Data consistency and redundancy is avoided
using normalization
It is comparatively less secure
Has a sophisticated data backup and recovery
system.
Some RDBMS systems are quite expensive for
smaller organisations.
Table 1.2: RDBMS System

Question 2.
Hierarchical database models could be explained as a model of organising in database into multiple
one to many relationships on the premise of single inheritance (one parent allowed multiple kids but
kids restricted to one parent).
Below is a table enumerating some of the advantages and disadvantages of this model?
Advantages Disadvantages
Easy addition and removal of new information Not suitable with more sophisticated
relationships e.g. many too many.
Data at the top of the hierarchy is very fast to
access.
The model tends to contain repeated data.
It functions best with natural hierarchies e.g.
assembly plants.
Data search is cumbersome and time consuming.
Table 1.3: Hierarchical Database
Network database model presents a flexible way of representing objects (Entities) and their
relationships. If is structure is viewed as a graph, its objects would equate to the nodes on the graph
while the arcs represents their relationships. It has the ability to appropriately cater to many to many
relationships. The table below shows some of its advantages and disadvantages.
Advantages Disadvantages
It is simple and easy to design Its structure could become complex owing to its
utilization of pointers
Can handle more entity relationship types Data manipulation requires a large number of
pointer adjustment
Its structure allows for easy data access and also
data integrity.
Implementing structural changes to a database is
very difficult.
Table 1.4: Network Database

Relational database models represent data in a series of tables where each table represents a
particular entity. Each individual table represents its attributes in columns. A single row in a
database (relational) table stand for a record. This model has the ability to cater to all kinds of
relationship.
Advantages Disadvantages
The structuring of the information in columns
and rows in respective tables makes them easy
to relate with and use.
Its hardware needs in terms of computing and
storage are quite large.
It is very flexible to use. Its ease of use could lead to poorly designed
systems.
Security control and authorization can be
implemented more easily.
Its definition and manipulation language isn’t a
perfect fit for object oriented programming
languages.
Table 1.5: Relational Database
Question 3.
Strengths and Weaknesses of the bottom up approach to Database design and Development: The
bottom up approach to design and development (database) ensures that more often than not all the
attributes and characteristics of the envisaged data structures are identified as the basis for the
database structure as its primary components but leaves the establishment of their interrelationships
and housing entities tedious to build especially with large data structures.
Strengths and Weaknesses of the top down approach to database design and Development: The top
down approach allows you to more often than not properly build the holistic data structure of the
system and correctly identify entities and the interrelationships but sometimes the respective
attributes of each individual entity might not be fully captured at first.
Justification of the usage of the top down approach in this project. In this project (AutoSeller) we
would using the top down approach to design and develop the database as it will afford us the
opportunity to properly envisage the database structure and correctly identify the respective entities
and their relationships. We could also afford the luxury of minor corrections of the constituent
attributes of the respective entities as the entities themselves are not sizeable.

TASK 2
Question 1.
In identifying all the possible entities in our problem scenario we would work with the premise that
all nouns in the problem statement are likely entities and all verbs are likely relationships, we can
then narrow down from there.
Having employed that technique below is a list of the identified entities in alphabetical order.
1. Buyers
2. Cars
3. Sales
4. Sellers
Question 2.
Below is a Diagram of the Conceptual Data Model.
Figure 2.1: Conceptual Data Model

TASK 3
Question 1.
Normalization: Normalization could be defined simply as a set of techniques applied to a related set
of database tables with the main aim of reducing the dependency and redundancy of data.
Below now is a guided description of steps take to achieve normal form in the AutoSellers scenario.
1. 1NF (First Normal form)
There are two major rules in the first normal form technique.
a. Each Table cell should contain a single value
b. Each record should be unique.
Test Case
The below table shows sales information for each seller over a short period in Autosellers records
Seller Name Seller Email Cars Sold Names of Buyers
Ritu [e-mail] Tata, Suzuki, Honda Modi, Khan, Raul
Rana [e-mail] Toyota, Nissan Sonia, Girish
Table 3.1: Seller Records
Applying the first normal form principles to this sample table for singularity of values and
uniqueness of record we can get this below table.
Seller Name Seller Email Cars Sold Names of Buyers
Ritu [e-mail] Tata Modi
Ritu [e-mail] Suzuki Khan
Ritu [e-mail] Honda Raul
Rana [e-mail] Toyota Sonia
Rana [e-mail] Nissan Girish
Table 3.2: Seller Records 1NF
2. 2NF (Second Normal Form)
There two major rules for application of this technique.
a. Table must be in first normal form
b. The primary Key must comprise of a single attribute.
Apply this principles to our last table we get this table below.

Seller ID Seller Name Seller Email Cars Sold Name of Buyers
001 Ritu [e-mail] Tata Modi
001 Ritu [e-mail] Suzuki Khan
001 Ritu [e-mail] Honda Raul
002 Rana [e-mail] Toyota Sonia
002 Rana [e-mail] Nissan Girish
Table 3.3: Seller Records 2NF
3. 3NF (Third Normal Form)
Two rules namely
a. Table must be in second normal form
b. All transitive dependencies must be removed.
Implementing this we take into cognisance the fact that records like cars sold, cars themselves and
also buyers details are still at present all dependent on adding a seller. Creating independent tables
for all these other entities will result in the tables shown in the conceptual data model.
Question 2.
Below is the Entity Relational Diagram (ER) for the AutoSellers scenario.
Sellers CarsSells
Buyers
Buys

Question 3.
In mapping this ER diagram (model) above we would segment it a take it in stages
1. Cars and Buyers Relationship.
This is a many to one relationship as many cars could be bought by one buyer but one car can’t be
sold to multiple buyers. Also going by that the sells entity should in effect have an attribute
(column) identifying the buyer.
2. Sellers and Cars.
This is a many to many relationship as many sellers can sell many cars and vice versa. Going by
that the relationship becomes an entity in itself, the sales (referred to earlier as sells) entity housing
unique identifiers for buyers, sellers and cars.
This produces 4 database tables namely
1. Cars
2. Sellers
3. Buyers
4. Sales.

TASK 4
Question 1.
Below is the script for creating the database for the AutoSellers scenario
-- tables
-- Table: Buyers
CREATE TABLE Buyers (
BuyerID int NOT NULL,
Name varchar(60) NOT NULL,
PhoneNo varchar(10) NOT NULL,
Address varchar(200) NOT NULL,
Email varchar(50) NOT NULL,
CONSTRAINT Buyers_pk PRIMARY KEY (BuyerID)
);
-- Table: Cars
CREATE TABLE Cars (
CarID int NOT NULL,
CarName varchar(50) NOT NULL,
CarModel varchar(50) NOT NULL,
CarPrice int NOT NULL,
CONSTRAINT Cars_pk PRIMARY KEY (CarID)
);
-- Table: Sales
CREATE TABLE Sales (
SaleID int NOT NULL,
Amount int NOT NULL,
Date int NOT NULL,
Sellers_SellerID int NOT NULL,
Buyers_BuyerID int NOT NULL,
Cars_CarID int NOT NULL,
CONSTRAINT Sales_pk PRIMARY KEY (SaleID)
);

-- Table: Sellers
CREATE TABLE Sellers (
SellerID int NOT NULL,
SellerName varchar(50) NOT NULL,
SellerPhone varchar(10) NOT NULL,
Address varchar(200) NOT NULL,
Email varchar(50) NOT NULL,
CONSTRAINT Sellers_pk PRIMARY KEY (SellerID)
);
-- foreign keys
-- Reference: Sales_Buyers (table: Sales)
ALTER TABLE Sales ADD CONSTRAINT Sales_Buyers FOREIGN KEY Sales_Buyers
(Buyers_BuyerID)
REFERENCES Buyers (BuyerID);
-- Reference: Sales_Cars (table: Sales)
ALTER TABLE Sales ADD CONSTRAINT Sales_Cars FOREIGN KEY Sales_Cars (Cars_CarID)
REFERENCES Cars (CarID);
-- Reference: Sales_Sellers (table: Sales)
ALTER TABLE Sales ADD CONSTRAINT Sales_Sellers FOREIGN KEY Sales_Sellers
(Sellers_SellerID)
REFERENCES Sellers (SellerID);
-- End of file.
Question 2.
Populating various tables with data.
1. Buyers Table
INSERT INTO ‘Buyers’ VALUES (‘0001’, ‘Khan’, ‘91467321’, ‘Mumbai’, ‘[e-mail]’);
2. Sellers Table
INSERT INTO ‘Sellers’ VALUES (‘0001’, ‘Rana’, ‘91465421’, ‘Chennai’, ‘[e-mail]’);
3. Cars Table
INSERT INTO ‘Cars’ VALUES (‘0001’, ‘Honda’, ‘Accord’, ‘400000’);
4. Sales Table
INSERT INTO ‘Sales’ VALUES (‘0001’, ‘400000’, ‘2019-02-19’, ‘0001’, ‘0001’, ‘0001’);

Bibliography
Anon., n.d. FUNDAMENTALS OF DATABASE SYSTEMS. In: FUNDAMENTALS OF
DATABASE SYSTEMS. s.l.:PEARSON.
Watt, A., n.d. /chapter-12-normalization/. [Online]
Available at: https://opentextbc.ca/dbdesign01/chapter/chapter-12-normalization/