Guidelines for Database Design and Implementation

Task 1.1: Selection of the case upon which the database design and implementation is to be based

irst, you need to identify a suitable case study (your choice or a choice from the list provided in the appendix) from which to derive your database requirements. This may be a situation based on a company with which you are familiar or in which you are (or have been) employed, or may be one (or based on one) that you have read about within the trade or academic literature or identified from their web presence. You need to ensure that your business situation is suitably complex to provide you with at least four strong entities, and at least one specialisation: generalisation structure, yet suitably scoped so as to not require a huge quantity of resultant tables (i.e., normally no more than 15) and subsequent input of sample data for querying purposes. It must not be based on a library (video, book, CD or film) and not just solely about orders of products. It should not be one that is based on any previous examination scenarios for this module, nor any exercises given to you within your DS&D module lecture material and/or module pack. Once researched and identified, a written scenario needs to be produced that (a) provides relevant background information on the organisation (e.g., its purpose, its principal operations/structure, its products/services, its target markets, etc.), and (b) provides an overview of what operations a database would need to support. Your scenario will probably be no less than one side of A4, but no longer than three slides of A4.

Task 1.2: Provide a conceptual database design for your scenario & the list of enterprise rules being modelled.

The second task is to develop an EER Diagram that captures the detailed requirements for the database system that you identified within the scenario you wrote to satisfy Task 1.1.  The EER Diagram needs to show any weak and strong entities, the primary keys for strong entities, and any relationships between entities (including any generalisation: specialisation structures). *:* relationships must be decomposed, and any actual traps identified should be eliminated using appropriate methods. For each entity, there should be an associated written list of all the attributes that the entity possesses which are not written on the EER Diagram. Any assumptions made during conceptual database design (i.e., anything that you assume that is not written in your Task 1.1 scenario) should be listed.

As well as the conceptual database design, you also need to provide the exact list of enterprise rules that your EER Diagram is diagrammatically representing. (Every relationship will need at least one enterprise rule, depending on its multiplicity and degree. Each binary relationship will typically have two enterprise rules associated with it, for instance.)

Task 2.1: Provide a Logical Database Design for your scenario

From your conceptual database design, derive a corresponding set of well-normalised tables.  Remember to indicate all primary and foreign key fields for each of the tables using suitable and consistent notation. All key and any non-key attributes should be listed within each table.

Task 2.2: Create the tables using Oracle DBMS

You need to create all the tables that you identified within your logical database design. Make sure the appropriate fields are defined as key, and that other suitable data integrity rules are enforced. Each of your tables should contain your PNumber as part of the table name. E.g. if your user name is ‘mit10sf’, then if you needed a Car table then you would create a table ‘mit10sfCar’. (Hint: make sure you create the tables in an appropriate order – for instance, those that have foreign keys cannot be created first – why? Think about it!).

Task 2.3: Create the four most useful indexes on your tables

You need to create a total of FOUR appropriate indexes on the tables using the CREATE [UNIQUE] INDEX statement. (Hint: it may be useful to consider what queries you wish to perform in Task 2.5 first, and don’t forget that primary keys don’t need any user-defined indexes as these are provided automatically by Oracle). Write a short explanation as to why you decided to create each index. 

Task 2.4: Data Population

Populate your Oracle tables with some fictitious yet appropriate test data (about FIVE records per large table and TEN records per small table (or as many rows as is relevant) should be enough)

Task 2.5: SQL Query writing

Define and run SIX queries of your choice (but appropriate to the scenario). Each query should require TWO or more of the following querying facilities, (and all of these facilities should be used at least once in your set of queries) and should be properly justified as to why the query would be useful to your case study organisation:

a.Selection of particular table columns

b.Inner Join of at least 2 tables

c.Outer Join of at least 2 tables

d.Use of count and/or another similar mathematical expression

e.Use of a sorting/ordering facility

f.A condition using “<”, “>”, LIKE etc.

g.A condition using IN, NOT NULL, or similar.

h.sub-query