Custom «Database Design and Implementation» Essay Paper Sample

The process of designing a database requires a lot of information from the organization. One needs to understand both the operational requirements and business requirements of the organization to come up with a good database design. Then one needs the knowledge to enable them create the database using these details. The design involves using a development lifecycle.

The process of creating this system consists of several steps. The process begins with the coming up of data that is used in the organization, and then identifying how this data is related to each other. The future is put into consideration and future data needs have to be put into the process. Then the use and generation of the data is determined.

To understand this topic we have to first understand some of the phrases used in this area. The ‘key’ is a group of several columns which identify a record or data in a unique way. An attribute is a column which has been named which represents the property. A relation is the claimed table which shows the data. Data is represented in these tables called relations.

After the first step, the usage of the data has to be determined. The unique fields have to be identified. This is information collected from the users. The day to day processes of the data and what that data stands for has to be investigated. Next you have to look at how the data works, how is it obtained, what it require on day to day basis, how does it communicate with other data forms in the organization. Finally the movements of the data has to be collected, this means that one has to study the size currently and in future terms.

These are the points that one needs to begin with before settling down to come up with a database design. Your model should incorporate and suit all this details. The next process is to determine the type of database design required. There are three types of database design;

• Physical database design,
• Conceptual database design, and
• Logical database design.

Physical database design

The physical database design is meant to process data efficiently. Information gathered before the designing process is used. This information includes the relations of the data plus its size estimate. Attributes are also defined here. The process the data undergoes on day to day basis, from past to present. Then what is expected for response, recovery, time, backup, security, retention and integrity and requirements is implemented here. After assessing the information required creating this design, several decisions are made on issues which affect performance and integrity of the design. They include issues such as the indexes, data arrangement and storage format among other issues. It is important that these procedures are followed to eliminate occurrence of any problems. The data is used to create a graphic representation called an entity relationship diagram. The ERD is then constituted into a logical design. It is from the logical design that the physical database design is carved out of. Ounce the logical design is available; it is much easier to come up with a physical database design. Different objects are just made from the logical design and turned into physical database objects. Attributes are turned into columns, entities into tables and relationships into foreign and primary keys. The job doesn’t stop there; the design needs to be studied as time passes. This is to ensure that the data is continuously monitored to keep upgrading the system every once in a while, as the data amount and type keeps changing (Lightstone, 2007).

In conceptual database design, there are several methods of coming up with the design. The first simpler one is where there is an entity language. The typological property can be changed into an entity. It has a demerit in that it overreacts and many topological properties become included in the database. In the second method, the typological properties can be defined as an entity.  The entity becomes as a question, with each property becoming an entity of this type. The question is asked about each language from the database, and the value of the typological property in the same language becomes the answer (Garmany, 2005).

Logical database design

In the designing of logical design, we have several steps which are followed. Since the data to be used is already available, this is what we implement with.  The key that can be used to find all the data available has to be found. The part key dependencies have to be removed, so that all data can be used being dependent on the key. All non key dependencies have to be removed so that all data is dependent on nothing else but the key. This is how you come up with a logistic design.

For the best success rates, the factors which guide the process must be followed. The developer must work in totality with the users. The structured methodology of the process must be followed. The data must be the lead determinant in the designing of these models. Structural and integral considerations must be incorporated into the data models. Combined methods of concept building, normalization and techniques for validating transactions into the data modeling method must be incorporated.

Implementation

The implementation process analyses all the procedural steps used in the designs. The implementation process starts with the decision on which organization subdivision is to be served by the database. Then, the functions which are to be served by the database are defined. The existing and planned applications should be converted into database system. The proposal for the management team is prepared and forwarded. Then the users for the designing team are chosen, followed by the selection of DBA (database administrator). Regular meetings and planning for periodic reporting for team is set up. Database management system products are selected. The documentation should be in formal proposal requests. A DBMS (database management system) vendor is also chosen. An implementation plan and scheduling is done. It includes the identifying of files which are to be converted. Programs which are in the specified applications need to be identified. Programmer hours which are needed to modify application programs are also identified. User clerical support which is needed to verify data by use of conversion is estimated.

Then database design is put into motion. Information requirements must be completed in details, before identifying data requirements. Then data structure and design specifications which are complete have to be determined. Approval of the design specifications is given after the review. Training requirements are developed together with the training schedule. All programmers should be well versed in the use of Data Manipulation Language (DML). Installation and tests are then carried out next, which includes the coding of DMCL and the schema. Representative programs need to be modified for DBMS. Then the generation of the database, testing and debugging all need to be carried out. The tests performed are reviewed. A detailed conversion plan is developed, it involves making of individual program assignment with each program to be modified and files to be loaded. Conversions involved should be approved. Applications which are in existence should be converted through bringing up one application at a time. The database should be updated and regenerated.  Revised applications can now be approved as they are converted. The database can now be used for new programs and applications. The DBMS access statistics should be monitored so as to modify anything else that is required. The database should be reviewed frequently. Finally, evaluate how successful the designs have been (Garmany).