Data Analysis and Design: A Critical Comparison of Data Models and Database Technologies

This document explores the critical aspects of data models and database technologies, emphasizing the importance of choosing the right model for specific applications. It delves into the benefits and limitations of various database systems, including relational, graph-oriented, and time series databases. The document also provides a comprehensive guide to designing and implementing a relational database system, including normalization, entity-relationship diagrams, and SQL query language. Practical examples and illustrations are used to demonstrate the concepts and techniques discussed. The document concludes with a critical evaluation of the implemented database system, addressing verification, validation, and control mechanisms.

Data Analysis and Design
1

Task 1
A.C.1.1 Critically compare different data models and Schemas
Choose the right data model, or opt for the most relevant data management strategy.
Data has always been of great importance to businesses. However, the advent of the web has literally
exploded their volume - as the methods to treat them - so that the role of the database becomes
eminently strategic. Added to this is the emergence in recent years of new database architectures that
make their choice even more complex. Each database system has its own characteristics, which reflect
how designers have chosen to understand the different issues related to data administration. Each of
these decisions results from compromises whose primary objective is to focus on what matters most.
For example, should data consistency prevail over the availability of the database? Should we favour
vertical scalability of an instance or, on the contrary, give priority to horizontal scalability?
The easiest way to choose the right model is to determine the value or importance of the relationships
between your application's data. A reflection that calls for other questions:
- How many relationships are there in the data?
- What is the complexity of these relationships? In other words, how many types of relationships are
there? Do these relationships have properties or annotations?
- How often does the application query these relationships during a user request?
Depending on the answer to each of these questions, we will be able to judge how important relations
will be within a dataset. If your data has a multitude of relationship points, their value will probably be
higher. Nevertheless, these answers can be subjective. By looking at the complexity of the relationship -
and its pecuniary value for the business - it is possible to choose the data management model more
appropriately.
.
2

1.2 Critically discuss the benefits and limitations of different database technologies
A poor choice of the data model can also affect the development of software and applications, as it
would create unnecessary complications for developers. But the main difficulty is due to the complexity
of the code required for the data access layer.
In terms of software maintenance costs, the choice of the wrong model can add to the "technical debt"
contracted for everything to work properly and come to disturb the work of developers, who may be
forced to develop alternative solutions or program more code to make everything turn. Ultimately, this
can result in overruns of budget or time - not to mention the frustration generated or the need to
designate culprits.
For example, one might decide to use a relational database to integrate all the customer data into a
"360 client" application. Although nothing prohibits it technically, it would involve a long reflection on
data modelling. Not to mention that one would probably have to develop more code to execute any
recommendation request, however simple.
With a graph database, maintenance of business issues can be done separately from the application. We
can then compile a complete profile of each customer in which are listed his last purchases, the
products he has viewed, the recommendations he has written, the movements made on his shopping
cart and the items he has saved on his wish list. This data can then be used to identify trends and who
buys which products. A graph-oriented database will, therefore, be easier to maintain over time, but
also more likely to generate value for the company.
3

1.3 Analyze different approaches to database design
The use of more complex data models is to be avoided when it is not fully justified. If we save a set of
operation histories in a graph-oriented database, using a query language like Gremlin will be difficult and
overly complicated. The simpler databases offer more readability for their performance and often result
in lower costs. In addition, the mapping between the logical modelling of data and their physical
representation within the system is clearer. In more sophisticated models, such as relational or graph-
based databases, query execution is often performed by query optimizers and a multitude of index
structures. This causes a multiplication of data paths and configuration options while complicating the
execution of queries. If we do not have the use of the advanced relationship-building capabilities that
are so important to these data models, you'll be paying for these features for nothing, in terms of
performance or effort on the part of the professionals responsible for their deployment.
Find the best configuration for an entire application or service
It is even more important to choose the right type of database for your application or use case that any
mistake in this area can be extremely expensive. However, most applications will create data that can be
used in various ways by the company. This can lead to situations in which several data models must be
used depending on the use of the data. This would be particularly the case for an application that would
use data from connected devices and then analyze them to identify trends and estimate their value.
In the long run, the amount of data generated by connected devices tends to grow in size as they send
updates back to the central servers at regular intervals. Also, if your business has thousands of devices
that update their status every quarter of an hour, you'll need a solid time series database. To analyze
this dataset, however, depending on the method chosen for data mining, we may need a different
configuration in parallel.
Depending on the value that can be generated by exploiting the data, it may be necessary to use
different data models for the same service or application. This, therefore, assumes that the data is
moved or transferred as a stream to its final destination. However, each data model - and each data
management platform that accompanies it - will have to be supported, starting with the production
systems.
In the near future, this kind of "multimodel" databases will become more important. Gartner predicts
that “by 2017, all major operational database management systems will integrate multiple data models,
4

relational and NoSQL, within a single platform, “according to a research note by Nick Heudecker, Merv
Adrian and Etisham Zaidi. The objective, in this case, is to facilitate support for production systems,
while still choosing the most appropriate data model for the application.
Going forward, the value of each data model must be carefully considered from the start. Any wrong
choice in this area can add unnecessary costs and software development costs. However, as applications
increase their data consumption and incorporate production deployments whose architecture is based
on the value generated by the data, the need for the concurrent use of multiple data models will also
increase. Developers will have to learn to consider data processing in a variety of ways. Fortunately, they
will be able to rely on multimodel data management tools to do this.
5

TASK 2:
2.1 Design a relational database system to meet given requirements provided in the given case study.
Here, I have designed the database structure to develop the database for Dominican college. The
Dominican college is located in the UK in a different location so, it required to develop the database. To
develop the database there are five steps to follow them. The first step is to design the database by
designing the ER and normalization. These two designing process is done here:
Figure 1 ER design to develop the Dominican database
Normalisation:
The database should be normalized and it does not have any of the duplicate key or anomalies in the
Dominican college. To solve this problem, introduce a normalization method which is presented here:
1. First Normalisation: In normalization, the duplicate key and anomalies are removed
6

Figure 2 I normalisation
2. Second Normalisation
Figure 3 2 normal Form
3. Third Normalisation
7

Figure 4 3 Normalisation
8

When we directly build the tables of a database in a database management software (Oracle, SQL
Server, DB2, Access, MySQL, PostGre), we are exposed to two types of problem:
a. We do not always know in which table to place certain columns (for example, the delivery
address is put on the table of the customers or in the table of the orders?);
b. We have difficulty in predicting the intermediate junction tables, for example, the table of
interpretations which is indispensable between the tables of the films and the table of the
actors).
It is, therefore, necessary to resort to a preliminary design stage. The techniques presented here are
part of the Merise methodology (Method of Study and Computer Realization for Enterprise Systems)
developed in France in 1978 (Tardieu et al.), which allows, in particular, to design an information system
in a standardized and methodical way. The purpose of this course material is to introduce the entity-
associations schema section "Conceptual Data Model (CDM)", the relational schema (sections "Data
Logic Model (LDM)" and "Physical Data Model (PDM)" ) "), And explain the translation between the two
(sections" Translation of a CDM into an MLDR "and" Retro-design "). The construction of the entity-
association schema can be done by studying the functional dependencies ("Functional dependencies"
section) and taking into account a number of unavoidable conceptual extensions ("Complements"
section).
Unfortunately, the following are not discussed here: constraints, treatments, relational language and
project management. For all these important notions, as related to the design of information systems,
the reader is directed to (Akoka & Comyn-Wattiau), (Matheron) and (Nanci et al.). Object modelling is
also not part of the tools presented in this document.
9

2.2 Build a relational database system for the database design in A.C 2.1 using an appropriate
database software application.
To build database, tables and firm should be developed and forms should be developed to create the
user interface.
Relationship Diagram
Figure 5 Relationship Diagram
Tables design view:
Figure 6 administrator Table
10

Figure 7 Branch Name
Figure 8 Course Program design view
Figure 9 Lecturer table design view
Figure 10 Manager Table
11

Figure 11 Student Design view table
Figure 12 Unit design view
Datasheet view
The tables are designed to store the data in the Dominican college database:
Here, I have present the datasheet view of the tables with stored data
Figure 13 unit table
12

Figure 14 Student datasheet view
Figure 15 Manager datasheet view
Figure 16 Instructor datasheet view
Figure 17 Course datasheet
13

Figure 18 Branch table
Figure 19 Administrator
Conceptual Data Model (CDM)
Before thinking about the relational schema of an application, it is good to model the problem to treat
from a conceptual point of view and independently of the software used.
Entities-associations
The conceptual modelling that we propose in this document for a universe whose data we want to
store, leads to the development of a very popular type of schema, the entity-association schema.
Entities and associations
An entity is a homogeneous population of individuals. For example, the products or items sold by an
enterprise can be grouped together in the same items entity (Figure 1), because from one article to
another, the information does not change in nature (each time it is a name, unit price, etc.
On the other hand, articles and customers cannot be grouped together: their information is not
homogeneous (an article does not have an address and a customer does not have a unit price). It is,
14

therefore, necessary to reserve for the two distinct entities: the item entity and the client entity. An
association is a link that has a precise meaning between several entities. In our example, the ordering
association is an obvious link between item and customer entities, while the delivering association
establishes the semantic link between item and vendor entities.
In this schema, the clients and supplier entities are not linked directly, but indirectly, via the items
entity, which is quite natural.
Attributes and identifiers
An attribute is a property of an entity or an association.
Again in our example, the unit price is an attribute of the item entity, the last name is an attribute of the
customer entity, the ordered quantity is an attribute of the ordering association and the delivery date is
an attribute of the association deliver.
An entity and its attributes should deal with only one subject to ensure consistency of the model. In our
example, it is better not to put vendor information in the item entity but rather in a separate vendor
entity (and linked to the item entity via the deliver association).
Then each individual of an entity must be uniquely identifiable. This is why all entities must have an
attribute without duplicates (that is, not taking twice the same value). This is the identifier that is
highlighted on the diagram, by convention. The customer number is a typical identifier for the customer
entity. An entity has at least one attribute (its identifier); on the contrary, an association can be devoid
of the attribute.
Cardinalities
The cardinality of a link between an entity and an association specifies the minimum and the maximum
of times that an individual of the entity may be concerned by the association.
15

2.3 Apply a range of database tools and techniques to enhance the user interface to meet the given
case study requirements.
For design the user interface, I have design forms to insert the data into the tables and to delete and
update the data from the tables of the Dominican college database.
Figure 20 Dominican college database
16

Figure 21 structure of design database
17

Figure 22 Create a database
18

Figure 23 Administrator Form
Figure 24 Branch Form
Figure 25 Course Program form
19

Figure 26 Instructor Form
Figure 27 Manager Form
20

Figure 28 Student Form
Figure 29 Unit Form
Reports to show the details of the queries to the users: The reports are designed to present the records
which are searched by the query from the database.
21

Figure 30 Administrator Query
Figure 31 Course Program Query
22

Figure 32 report of query
Figure 33 Instructor Query report
23

Figure 34 Student Query Report
Figure 35 Unit Query Report
24

Tools like normalization and standardization are used to enhance the application interface. Entity
standardization (important): All entities that are replaceable by an association must be replaced.
Normalization of names: the name of an entity, association or attribute must be unique. For entities, use
a plural common name (for example: customers); for associations, to use an infinitive verb (for example:
to carry out, to concern) possibly to the passive form (to be ordered) and accompanied by an adverb (to
take place in, during, to); for the attributes, use a common name (for example: name, number,
description, description), possibly accompanied by the name of the entity or the association in which it is
located (for example customer name, number d Article). When the same name remains several times, it
is sometimes symptomatic of modelling that is not completed.
Two homogeneous entities can be merged. If two attributes contain the same information, then
redundancy not only leads to waste of space but also a great risk of inconsistency. Normalization of
identifiers: each entity must have an identifier. We should avoid identifiers composed of several
attributes (such as an identifier formed by name and surname attributes), because on the one hand it is
bad for the performances and on the other hand, the uniqueness supposed by such an approach ends
sooner or later to be denied; we prefer a short identifier to make the search as fast as possible (avoid
strings such as a license plate number, a social security number or a postal code); we also avoid
identifiers that may change over time (such as license plates or provisional social security numbers).
Conclusion: the identifier on an entity-association schema (and therefore the future primary key in the
relational schema) must be an integer, preferably incremented automatically. Attribute normalization
(important): replace multiple copies of attributes in an additional association of maximum cardinalities n
and not add computable attributes from other attributes. Indeed, on the one hand, the attributes in
several copies pose problems of scalability of the model and on the other hand, the Calculable attributes
induce a risk of inconsistency between the values of the basic attributes and those of the calculated
attributes.
25

Task 3
3.1 Explain the benefits of using manipulation and query tools in a relational database systems per the
requirements are given in the case study.
In this section, I have present the benefits of the data manipulation and explain the query tools which
are used to designed the database tables and query to store and search the data.
Data manipulation: the data manipulation is the process which is used to insert the data, update data,
delete the data and alter the data. With this process, we can plan the data insertion and its storing
space (Floyd, 2014).
Benefits of data manipulation
1. With this language, the data is specified the incorrect way
2. It also helps to create the relationship between the designed entities and tables of the Dominican
college.
3. The stored data is able to change the data manipulation statements
4. The data manipulation is working as to create the interaction between the users and the database
system by designing the user interface (Evert, 2015).
Query Tools: The query tool is specified to generate the queries and tables in the Dominican college. It
is mainly used to create the interaction between the customer, user and database system. There are
various query language that are given as DDL, DML and SQL etc. these query tools are used create the
database with user interface to make the interaction between users and system.
Benefits of the Query languages and tools: The query tool helps to store the data and help to access
the data from the Dominican College. It helps to manage the database and it not required any special
requirement to perform it. The query tools are also help to secure the database with providing security
features to the database (Techopedia, 2017).
Query tools help to translate the physical model into the relational database. In the context of relational
databases, the physical model must be converted into a normalized relational schema (by unravelling
any optimizations and renaming the columns of the tables to ensure the uniqueness and explicit
(encoded) nature of the names), then apply the translation rules in the in the opposite direction.
26

Step 1: Each table whose primary key does not contain a foreign key becomes an entity whose identifier
is the primary key of the table and whose attributes are the columns of the table that are not foreign
keys.
Step 2: Each table whose primary key is composed exclusively of foreign keys that reference several
primary keys, becomes an association around which all the maximum cardinalities are equal to n, that is
to say, a binary association of type n: m, either a ternary association or more (the other non-foreign key
columns of the table become attributes of the association).
Step 3: The remaining foreign key columns become 1: n binary associations, if there is no uniqueness or
1: 1 constraint, if there is a uniqueness constraint (it must find a name for this association).
Step 4: The minimum cardinality is 1 for foreign keys that are part of a primary key or have a (non-
empty) constraint, otherwise it is 0.
Query 1: To print the course name, unit name with course id, unit id (Tutorialspoint, 2017)
Figure 36 SQL view
Figure 37 Output
Query 2:
Figure 38 SQL view
27

Figure 39 Output of Query 2
Query 3:
Figure 40 SQL view
Figure 41 Output Query3
Query 4:
Figure 42 SQL View
Figure 43 Output for query 4
Query 5
28

Figure 44 Query 5
Figure 45 Output of Query 5
29

3.2 Implement a query language into the relational database system that has been implemented in
task 2.2
With the help of query language, I have designed the database queries which are given below:
Query 1: To print the course name, a unit name with course id, unit id (Tutorialspoint, 2017)
Figure 46 SQL view
Figure 47 Output
Query 2:
Figure 48 SQL view
Figure 49 Output of Query 2
30

Query 3:
Figure 50 SQL view
Figure 51 Output Query3
Query 4:
Figure 52 SQL View
Figure 53 Output for query 4
Query 5
Figure 54 Query 5
31

Figure 55 Output of Query 5
The query tools help in the translation of physical model to the actual database management system.
Unfortunately, these four steps are not enough to translate all possible relational schemas. In particular,
the tables require the insertion of additional steps. Each table whose primary key consists exclusively of
foreign keys that reference a single primary key becomes a sub-entity or a sub-association (the other
non-foreign non-key columns in the table become attributes of that sub-entity).
Each table whose primary key is partially composed of foreign keys comes either from an optimization
that it undoes (as in Figure 35) or from a relative identifier of an entity as in the section "Relative
identifier or link identifier" (in which case the other non-foreign columns of the table become attributes
of this entity). No complete situation, or almost, can be perfectly modelled if the designer is satisfied
with the functionalities approached at this stage of the document. If only to understand the tables, it is
necessary to introduce new notations on the entity-association schema. The three major extensions
presented in this section are part of version 2 of Merise (Planet, et al.). They make it possible to deal
with more realistic situations and often in a simpler way. In this section, we take the approach of
studying the direct functional dependencies on the minimum coverage graph, then translate this graph
into entity-association schema, to finally obtain a relational schema. The concepts discussed here no
longer allow the relational schema to be written verbatim unambiguously. In order to remove any
ambiguity as to which primary key is referenced by such foreign key, it is imperative to represent the
relational schema graphically, which we are content to do.
Aggregation
An association is not necessarily established exclusively between entities.
Association of type 1: n
32

Consider the example of Dominican College. The functional dependence n ° student + n ° course? n °
teacher is the first non-elementary functional dependency to an identifier that we encounter. This type
of functional dependence encourages us to create a type 1: n binary association between the student
entity and the n: m type binary association between the course and teacher entities.
33

3.3 Critically evaluate how meaningful data has been extracted through the use of query tools for the
given case study.
The database is the collection of the data which is required to store about the students, lecturer, unit,
course, campus and manager who manages the course allocation to the students and tutor to teach. The
database is implementing to store the meaningful data with using the SQL queries and provide the
appropriate details (Slamecka, 2017).
SQL advantages to develop the database:
 The SQL is mostly used a query language to develop the database.
 The SQL query language is the very interactive language that helps to find the specific and
appropriate information.
 It independent language and portable, it executes on any type of devices like laptop, desktop
and tablets or mobile devices.
The SQL is used to extract the data by design the table, and query. The most important thing is that it
can be applicable in MS access that makes easy implementation (Law, 2004).
Tables that are designed to store the database:
Figure 56 unit table
Figure 57 Student datasheet view
34

Figure 58 Manager datasheet view
Figure 59 Instructor datasheet view
Figure 60 Course datasheet
Figure 61 Branch table
35

Figure 62 Administrator
Meaningful data is extracted for the purpose of optimization. Optimization of performance in computing
time is always to the detriment of the consumed memory space. In the worst case, reducing the
response time consists in deliberately de-standardizing the information system, with all the risks of
inconsistency and the management problems that entails. For relational databases, optimization that
aims to speed up queries can go through: adding indexes to tables (at least on primary key columns and
foreign keys); these indexes consume additional memory space, but the database remains normalized;
adding calculated columns or some redundancies to avoid costly joins (in which case the database is de-
normalized); it is then necessary to ensure that the consistency between the columns is respected either
by the use of triggers or in the client applications of the information system; the removal of the
constraints of uniqueness, non-emptiness or foreign key (in which case the integrity of the data must be
ensured by the customer code of the information system). For example, the orders table can be deleted
and the order date is added to the order line table. So we give up the third normal form since the order
date is repeated as many times as there are lines in the command, but we avoid an expensive join in
computation time during SQL queries.
The most valuable advice, in terms of optimization, is to never optimize a priori, but always a posteriori,
that is to say in response to a slowness that the RDBMS is not able to solve alone. It is then necessary to
measure the gain of any manual optimization by performing tests (before/after timing) on a significant
volume of data and preferably in operation
36

Task 4
4.1 Critically review and test the implemented relational database system
In most cases, the job of the database designer is not to create a database from scratch, but rather to
correct or extend an existing database. In this case, for initial work is a physical model and the method
of reverse engineering or reverse engineering is to translate the physical in a conceptual model, change
the data received then change the physical model accordingly. Thus, testing can be achieved by doing
this reverse engineering. For testing, triggers programmed in the final database could be used. The
triggers are SQL procedures which, in our example, would allow each insertion or update of rows in the
betting table, to ensure that a bet cannot involve a student in a course in which he does not participate.
However, there is a simpler solution that relies solely on referential integrity. These integrities are inbuilt
and can be used in testing.
In reality, the true direct functional dependence is no student + no course) + no teacher. The fact that a
ternary association (or more) disappears in favour of one or more aggregations is very common when
modelling a complete situation. So much so that we can assume that an entity-association schema
without aggregation is usually false. In our example, the translation of the new functional dependency
into an association of type n: m is done by applying, as usual, step 4 of the section "Translation to a
schema entities-associations ".
Query 1: To print the course name, a unit name with course id, unit id (Tutorialspoint, 2017)
Figure 63 SQL view
Figure 64 Output
37

Query 2:
Figure 65 SQL view
Figure 66 Output of Query 2
Query 3:
Figure 67 SQL view
Figure 68 Output Query3
Query 4:
Figure 69 SQL View
38

Figure 70 Output for query 4
Query 5
Figure 71 Query 5
Figure 72 Output of Query 5
39

4.2 Create documentation to support the implementation and testing of the relational
database system that has been implemented in Task 2.2
General prerequisites
Here are the main elements to prepare before installing Use it Flow (to be defined according to the
desired configuration): Subscription and installation of internet access, Subscription and installation of a
mailbox or RVA (Atlas 400, GXS, ...), Allow management software to generate incoming / outgoing
information flows to communicate with Use it Flow, Define the general conditions of the electronic
exchanges between the partners (interchange agreement), To regulate the administrative formalities
concerning the formations (Prologue is a training body), Establishment of a backup device allowing an
alternative mode of communication with partners in case of breakdown, Establishment of the technical
means to perform the tests while another system is already in production (secondary server, secondary
line, etc.), Implementation of the technical prerequisites (see below):
Technical prerequisites
Hardware concerning the server Hardware (server) Minimum configuration recommended * Processor
2.5 GHz processor 3 GHz processor or superior Resolution (screen size) 1280 x 1024 (19 inches)
1920x1080 (22 inches) RAM 4GB 8GB Available disk space 2 GB for the software 10 GB for data 4 GB for
the software 15 GB for the data (depending on the treatments implemented). Digital Prologue advice to
set up a system of scheduled backup and archiving after installation.
The Use it Flow solution can be used in a virtualized environment "VM-Ware, Hyper-V ", when there is
no device used with the solution. In case the solution is installed in a virtualized environment, the user
license agreement must be followed. The end user is forbidden to put the software and its
documentation at the disposal of third parties, directly or indirectly, for whatever reason. [...] The end
user is prohibited from reproducing, in whole or in part, the software and/or the related documentation.
40

4.3 Create user documentation for the developed relational database system in task
2.2
Software concerning the server:
The following table outlines the minimum requirements for the proper operation, as well as the
recommended versions. Except for database systems like Microsoft SQL Server (excluding Express
version), the installation of these software, in case they are not present on the server, are required. The
client must ensure there are no other versions of this software on the server. Operating System
Microsoft Windows 2008 SP2 and 2012 SP1. In 32 or 64 bits of Microsoft Windows 7 SP1 and 8.1. In 32
or 64 bits Debian 7.x (Wheezy) in 64 bits Base of data Use it Flow requires at least one version of the
following DBMS: MySQL Pro 5.1.x licensed MariaDB 10.x or 5.5.x Microsoft SQL Server (2008 or 2012,
Standard or Express can be used) HTTP server We recommend Apache 2.4.x (2.2.x supported) : The user
under which the Apache server is running must have the rights to read / write to directories containing:
Use it Flow website the Use it Flow application (engine storage directory) We recommend PHP 5.5.x
(5.4.x supported) in Apache module with the following PHP modules (from the PHP distribution and
from activating during installation): XML XSL MySQL CURL DOM GD 2 mbstring LDAP SOAP If the DBMS is
Microsoft SQL Server, the following modules should to be installed: PDO PDO_ODBC
Tables design view:
Figure 73 administrator Table
41

Figure 74 Branch Name
Figure 75 Course Program design view
Figure 76 Lecturer table design view
Figure 77 Manager Table
42

Figure 78 Student Design view table
Figure 79 Unit design view
Datasheet view
The tables are designed to store the data in the Dominican college database:
Here, I have present the datasheet view of the tables with stored data
Figure 80 unit table
43

Figure 81 Student datasheet view
Figure 82 Manager datasheet view
Figure 83 Instructor datasheet view
Figure 84 Course datasheet
44

Figure 85 Branch table
Figure 86 Administrator
Forms
45

Figure 87 Dominican college database
46

Figure 88 structure of design database
47

Figure 89 Create a database
48

Figure 90 Administrator Form
Figure 91 Branch Form
Figure 92 Course Program form
49

Figure 93 Instructor Form
Figure 94 Manager Form
50

Figure 95 Student Form
Figure 96 Unit Form
51

Reports
Figure 97 Administrator Query
Figure 98 Course Program Query
52

Figure 99 report of query
Figure 100 Instructor Query report
53

Figure 101 Student Query Report
Figure 102 Unit Query Report
54

4.4 Explain how verification and validation has been addressed in the developed
relational database system.
For verification, we must ensure for the day of the installation that Application will be able to access to
the server on which the database system will be installed over the Internet (in using remote control
software such as " Team viewer "). If this installation is not possible, we must ask system admin for an
installation on site (this request will lead to the establishment of another Technical Proposal and
Financials)
Query 1: To print the course name, a unit name with course id, unit id (Tutorialspoint, 2017)
Figure 103 SQL view
Figure 104 Output
Query 2:
Figure 105 SQL view
55

Figure 106 Output of Query 2
Query 3:
Figure 107 SQL view
Figure 108 Output Query3
Query 4:
Figure 109 SQL View
Figure 110 Output for query 4
Query 5
56

Figure 111 Query 5
Figure 112 Output of Query 5
.4.5 Explain how control mechanisms have been used in the developed relational
database system.
Control mechanisms are implemented through the ODBC drivers which are interfaced with all the
languages used for the two types of system environments (Windows, Linux). It's this layer of
communication that is used for Microsoft SQL Server access except for MySQL and MariaDB. In the PHP
configuration, the PDO and PDO_ODBC modules are used to PHP access to ODBC in Windows or Linux
environment. They are provided with PHP, and should just be enabled, if not. The presence of these
modules is tested at the "prerequisite" option of the procedure installation of Use it Flow. If not install in
the case of Linux, the package: php5-odbc: ODBC module for php5.
57

References
[Panet et al. ] Panet, G., Letouche, R. and Tardieu, H. Merise / 2: Advanced Merise Models and
Techniques. Organizational Edition, 1994. This book describes version 2 of the method.
[Tardieu et al. Tardieu, H., Rochfeld, A. and Coletti, R. The Merise Method. Principles and tools.
Organizational Edition, 1986. This is the reference book by the authors of the method.
Akoka, J. and Comyn-Wattiau I. Relational Database Design. Vuibert Informatique. This very didactic
book contains good exercises on the design phase of an information system.
Evert, S., 2015. An NLP approach to the evaluation of Web Corpora. Guest lecture, Leuven, 17.
Floyd, A. and Du, H., 2014. EDNA, a Software Tool for Verifying the Normalisation of Relations during the
Logical Database Design Process. Friday 4th July 2014 Southampton Solent University, p.51.
Gabay, J. Learn and practice Merise. Masson, 1989. This very synthetic book allows you to practice the
method.
Law, Y.N., Wang, H. and Zaniolo, C., 2004, August. Query languages and data models for database
sequences and data streams. In Proceedings of the Thirtieth international conference on Very large data
bases-Volume 30 (pp. 492-503). VLDB Endowment.
Matheron, J.-P. Understanding Merise. Eyrolles, 1994. This very accessible book really helps to
understand the method.
Nanci, D., Espinasse, B., Cohen, B. and Heckenroth, H. Information Systems Engineering with Merise.
Sybex, 1992. This comprehensive book details the method as a whole.
Slamecka, V. 2017. Information processing. [Online] Encyclopedia Britannica. Available at:
https://www.britannica.com/technology/information-processing#ref212132 [Accessed 10 Mar. 2018].
Techopedia, 2017. What is a Logical Data Model (LDM)? - Definition from Techopedia. [Online]
Techopedia.com. Available at:https://www.techopedia.com/definition/30599/logical-data-model-ldm
[Accessed 10 Mar. 2018].
58

Tutorialspoint, 2017. MS Access Create Tables. [Online] www.tutorialspoint.com. Available at:
https://www.tutorialspoint.com/ms_access/ms_access_create_tables.htm [Accessed 10 Mar. 2017].
59