Let’s talk about data! Previously we discussed the Database Development Process Steps. This topic is in high demand, which is not surprising considering the ever growing importance of data in all processes taking place in the world. Data collection, systematization, protection and security – database development process includes all of these touchpoints. Whether you will take advantage of the database that fully satisfies your needs, or not, depends on how specifically you create your database and which tools you choose. For this reason, today we’d like to go deeper in the database realm and elaborate on Database Development Methodologies. Stay tuned!
What Is a Database Development Methodology?
When creating a database developers need to go through a number of subsequent stages. These stages constitute a methodology for developing a database, making up the most complete description of technologies for the presentation, storage and processing of data by computer systems.
The development methodology is understood as the process of describing the system under development, guided by a finite number of certain stages, laws and rules which are applied for the development of individual components and notations that determine the formats for the representation of the developed elements.
Development of databases is also performed using specific development methodologies that include several stages, from analyzing the domain and related documents to describing the procedures for processing and presenting data. The database development is based on a preliminary analysis of the subject, on defined main and additional functions, as well as on a description of the environment of each function. This information gives a developer a rationale for using the necessary entities when modeling and representing the necessary attributes in them, organizing the correct relationships between those entities.
In general, a database development methodology defines:
- Design process;
- Methodology for calculations and criteria for evaluating alternative solutions at each design stage;
- Information requirements as an input to the development process;
- Means for describing the initial data and displaying the results of each design stage.
Database Development Methods
The database development methodology provides for the division of the entire development process into several phases, each of which consists of several stages.
Generally, the database design methodology is divided into 3 main phases:
- Conceptual design, which consists in the creation of a conceptual representation of the database, including the definition of the types of the most important entities and the relationships between them.
- Logical design, which lies in transforming a conceptual representation into a logical database structure, including designing relationships.
- Physical design, which implies deciding how the logical model will be physically implemented (using tables) in the database created using the selected DBMS.
We will provide you with an overview of the database design methodologies, including a brief description of the main steps involved in each of the phases mentioned above.
Conceptual Database Design
The conceptual database design comes down to constructing a description of the domain in terms of a formal language, for example, in terms of entities and relationships model.
A conceptual model is a certain set of concepts and connections between them, which form the semantic structure of the subject area. This model is the initial prototype of the future database. Basically, here an abstract model is created.
The conceptual database design is totally independent from such elements as the type of the DBMS, the set of application programs to be created, the programming languages used, the type of the chosen computing platform, as well as any other aspects of physical implementation.
During development, the conceptual database model is constantly tested and verified to meet user requirements. The finished conceptual database model is the source of information for the logical database design stage.
To build a conceptual model, you need to have a good knowledge of the subject area, its semantics, as well as understanding of the logical relationships of its information.
Here are the steps taken in terms of the conceptual database design:
- Definition of entity types;
- Determination of the types of links;
- Definition of attributes and linking them to types of entities and relationships;
- Specification of attribute domains;
- Definition of the attributes that are potential and primary keys;
- Creation of “entity – relationship” diagrams;
- Discussion of the local conceptual model with the end user.
As a result of the conceptual design phase we get:
- A formal description of software information support;
- A detailed and strict description of data stores;
- A detailed description of data streams;
- A detailed description of the hierarchy and specification of tasks that are being solved;
- A detailed description of the rules and restrictions in the area;
- A description of information objects or concepts of the subject area and the relationships between them;
- A description of the integrity constraints, namely, the requirements for the admissible data values and for the relationships between them.
Here are the key success factors:
- Deep understanding and practical experience of using languages for describing the conceptual model;
- Knowledge of methods for designing a relational model or other data models.
Logical Database Design
Logical design is the creation of a database schema based on a specific data model. With the help of the logical database design methodology the high-level data representation is transformed into the structure of the used DBMS. The main goal of this stage is to eliminate data redundancy using special rules called “normalization”. This minimizes data repetition and possible structural changes to the database during update procedures. This is achieved by dividing (decomposing) one table into two or more, and then using the navigation operation in queries. Note that navigational search slows down the performance of the database, i.e. increases the response time to a query. The resulting logical structure of the database can be quantified using various characteristics (the number of references to logical records, the amount of data in each application, the total amount of data). Based on these assessments, the logical framework can be improved to achieve greater efficiency.
At the logical design phase, the specificity of a certain data model is taken into account, but the specificity of a specific DBMS may not be taken into account.
Usually, the logical database design phase (for a relational DBMS) includes the following activities:
1.Building and validating a local logical data model:
- transformation of a local conceptual model into a local logical model;
- definition of sets of relations based on the structures of the local logical data model;
- model validation using normalization rules;
- model validation in relation to the user’s transaction;
- creation of the “entity – relationship” diagrams;
- definition the requirements for maintaining data integrity;
- discussion of the local logical model with the end user.
2. Creation and validation of the global logical data model:
- merging local and logical models into a single model;
- checking the global logical model;
- checking if the problem can be extended in the future;
- creation of the final version of the “entity – relationship” diagram;
- discussion of the global logical model with the end user.
The logical database model is the source of information for the physical design phase. It provides a developer of the physical database model with the means to perform a comprehensive analysis of various aspects of working with data, which is critical for choosing an effective design solution.
Physical Database Design
The physical database design is the process of creating a description of a specific implementation of a secondary storage database. It provides a description of the data storage structure and access methods designed to ensure the most efficient access to information.
Basically, the physical database design phase results in the creation of a database schema for a specific DBMS. The specificity of a certain DBMS may include restrictions on naming database objects, restrictions on supported data types, etc. In addition, the specificity of a certain DBMS in physical design includes the choice of solutions related to the physical storage environment (choice of methods for managing disk memory, dividing the database by files and devices, data access methods), creating indexes, etc.
Generally, at the physical design phase, issues related to system performance are resolved, data storage structures and access methods are determined.
The physical design phase of the database provides for the developer to make the final decision on how to implement the database under development. Therefore, physical design is necessarily carried out taking into account all the features of the used DBMS. There is always a connection between the physical and logical design phases, since decisions made at the stage of physical design in order to improve the performance of the system may require the revision of the logical data model.
The physical database design phase (for a relational DBMS) comes down to the following steps:
1. Transfer of the global logical data model to the target DBMS environment:
- creation of the main tables in the DBMS environment;
- implementation of enterprise business rules within DBMS.
2. Designing a physical representation of a database:
- analysis of transactions;
- choice of file structure;
- definition of secondary indices;
- data redundancy control;
- determining disk storage requirements.
3. Development of protection mechanisms:
- development of custom representations;
- determination of data access rights.
Typically, the results of the logical and physical database design phases are the following:
- System architecture;
- System module diagrams;
- Logical and physical circuits;
- A database and file schema;
- Detailed time and capacitive characteristics;
- Software specifications;
- Manual procedure specifications;
- A draft user manual;
- An agreed implementation strategy, which is made up of plans for the acceptance and delivery of the system, organizational preparation, measures for collecting data, switching to a new system and installing equipment;
- A system test plan;
- A draft version of operational documentation;
- A refined system development plan.
All in all, as a result of this stage, a project that ensures the satisfaction of application requirements, taking into account the existing technical constraints should be created. The key success factors in achieving this goal are:
- Knowledge of hardware and software capabilities;
- Understanding of applied needs;
- Making informed compromise decisions;
- Identifying and tackling potential problems.
Differences Between Database Design Methods
The difference in data presentation levels at each design stage is presented in the following table:
| Conceptual Design | Logical Design | Physical Design |
| Entities
Attributes Relationships |
Records
Data items Relationships between records |
Data grouping
Indices Access methods |
| Analyst’s view | Programmer’s view | Administrator’s view |
Critical Factors for Successful Design and Development of Databases
Finally, here are some guidelines to successfully complete your database development:
- Maintain constant and active communication with future users of the application;
- When conducting data modeling procedures, follow the guidelines provided in the discussion of the proposed methodology;
- Design the system based on existing data characteristics;
- Complement the procedures suggested by the methodology with technological methods for conceptualizing, normalizing, and verifying the integrity of transactions;
- Use charts to represent your data model as widely as possible;
- Use Database Design Language (DBDL) tools to describe additional semantic data requirements;
- In addition to data model diagrams, develop a data description vocabulary.
- Do not hesitate to return to the steps you have already completed if required for optimal results.
Database Development Methodologies by JoinSoft
JoinSoft is a trusted Custom Software Development Services provider that offers tailored solutions to take your business to the next level.
Our Database Solutions are developed with regard to your specific needs, as well as time and budget considerations.
We offer much more than a spreadsheet solution. We create innovative database software for the most efficient and convenient access to information. Our specialists use the leafing frameworks to create reliable and secure database solutions.
Flexibility and constant feedback lie in the core of our client interaction. We always make sure our clients are aware of the status of their project. Always open for comments, we are ready to ongoingly tailor our products to skyrocket clients’ satisfaction.
Unleash the power of Big Data for your business with JoinSoft!
Final Thoughts About Database Development Methodologies
Congratulations, now you know what database designs are, as well as how and when to apply them to extract maximum advantage from their operation. Note, that every methodology should be regarded from the standpoint of the specific project in its uniqueness and entirety.
JoinSoft offers professional Custom Software Development services to help you organize the collection and use of data in the most efficient and user-friendly way. If you are looking for a database solution for your business, don’t hesitate to contact us!
