PARTITIONING DATABASE
The major problem of the database occurs in system performance. System performance is the region where full scanning of big relational tables occurs involving an unusual amount of time on query processing. Horizontal, vertical partitioning or combining both solves the problem of the relational table (Alsultanny, 2010).
Vertical partitioning of the database
The database system often accesses substantial data amounts so that it can update or retrieve a proportionally small amount of values suggested by the user, as a result of the greatness of the row length average compared to the extracted data or row modified. For instance, projecting individual characteristic of the relational table, having a schema comprising of many features having values occupying a significant amount of space (Alsultanny, Y. 2010).
Delegate your assignment to our experts and they will do the rest.
Vertical database partitioning separates relational table into different pieces called partitions and copies primary key into each partition. Partitioning helps in reduction of the average length of the row and as a result, minimizing write and read operation total number (Alsultanny, 2010).
Horizontal partitioning of the database
Horizontal portioning involves breaking the relational table into small pieces, referred as portions, in order of the values range of a given feature or characteristic amalgamation referred as the partition key. The horizontal database partitioning minimizes the read operations total number required to enter into the rows which depends on the partition key value given in the query (Alsultanny, 2010).
NORMALIZATION
Database normalization is the traditional method of elimination redundancy of data and the irregularities it produces.
Normalization process involves examination of multi-valued, functional and project join dependency occurs to create a relation into several relations. Normalization results to a set of relationships which meets the necessity of the distinct normal form levels. The higher the normal from the level reached in the database, the lower the chances of irregularities of data occurrence (Press, R. T. I. 2011)
Vertical decomposition in normalization
Relational database design depends on vertical disintegration to make relations regular and prevent redundancy. The model decomposes relations into little ties. After vertical disintegration, the combination of two normalized form occurs hence data retrieval (Press, 2011)
Vertical decomposition eliminates data redundancy and removes the problem of the first potential anomaly of data. Therefore, decomposition helps to improve the quality of data. Decomposition relies on functional dependency, which declares that in a relationship one field's value determines another field's value of a particular row (Press, 2011)
Vertical decomposition process normalizes relation model into distinct normal form levels like 4th usual form, Boyce-Codd and the 5th normal form. The types are the highest level and are the objective of the process of database normalization (Press, 2011)
Change in relation design
After normalizing data into 5NF or BCNF by vertical decomposition, further steps taken to stop data anomalies includes Field level disentanglement and horizontal decomposition.
Field-Level Disentanglement
Field level disentanglement untangles interrelationships of data occurring at field level. The approach is conducted by further analyzing of the relation. Removal of restricted domain occurs through changing the relationship hence use of disjoint subsets. Using disjoint subsets involves replacing the field. The new design stops data anomalies. Field level disentanglement is vital in that the changes occur at the level of design (Press, R. T. I. 2011)
Horizontal Decomposition
The second option is called horizontal decomposition which stops the restricted domains at the level of design. The approach decomposes the relations horizontally by relation splitting into severalties having a particular table structure. The objective is to eliminate restricted domains occurring as a result of the limitation in the number of subset domain.
Importance of Improving Database Design
Improving database design is crucial in that it assists in minimizing resources consumption like the space of a hard disk and reduces the time used to retrieve data. Data organization method applies to the area of transmitting and storing data. Most of the applications processing data demand enormous data volume storage, and the quantity of this applications continually keeps rising as computer usage expands to other areas (Alsultanny, 2010).
Partitioning methods both vertical and horizontal are essential in speeding retrieval database retrieval (Alsultanny, 2010).
RAID technology level 1 with vertical partitioning combination, helps in improving the retrieval of the database and keeps reliability (Alsultanny, 2010)
Normalization helps to give integrity and maintain the database, enhances the quality of data and data flexibility (Teorey, 1999), minimizes redundancy of data and improves the consistency of data by decreasing irregularities. And also normalization achieves behavioral and structural advantages in the design.
Impact of Database Design
Database design improvement reduces the cost of production levels and programming levels (Press, R. T. I. 2011)
In conclusion, improving the database design involves partitioning and normalization. The importance of partitioning and normalization includes giving integrity and maintaining database, improving retrieval of database and minimizes consumption of hard disk space hence saving the costs of production and programming levels.
References
Albarak, M., Alrazgan, M., & Bahsoon, R. (2017). Database Normalization Debt : A Debt-Aware Approach to Reason about Normalization Decisions in Database Design. arXiv preprint arXiv:1711.06109 .
Alsultanny, Y. (2010). Database management and partitioning to improve database processing performance. Journal of Database Marketing & Customer Strategy Management , 17 (3-4), 271-276.
Press, R. T. I. (2011). Improving Data Quality in Relational Databases: Overcoming Functional Entanglements.
Teorey, T. J. (1999). Database modeling & design . Morgan Kaufmann.
